Aneuploidy compensatory mechanisms and genome-wide regulation of gene expression in Drosophila melanogaster

Lina Lundberg

Department of Molecular Biology

Umeå University 2013, Sweden

This work is protected by the Swedish Copyright Legislation (Act 1960:729) ISBN: 978-91-7459-659-5 Cover photo by: Glenn Landgren

Back side photo: Section from an expression array (enlarged)

Electronic version is available at http://umu.diva-portal.org/

Printed by: Print & Media Umeå, Sweden 2013

“The bad news is time flies. The good news is you’re the pilot!”

~Michael Althsuler

Till Faster Kicki

Table of CONTENTS

i

TABLE OF CONTENTS

TABLE OF CONTENTS ....................................................................... i

LIST OF PUBLICATIONS ...................................................................iv

TERMINOLOGY AND ABBREVATIONS .............................................. v

ABSTRACT ..................................................................................... vii

SVENSK SAMMANFATTNING ........................................................ viii

INTRODUCTION ............................................................................. 1

EPIGENETICS ....................................................................................... 1 Cell differentiation ......................................................................................... 2

CHROMATIN ....................................................................................... 2 Histone acetylation ............................................................................................ 5

H4K16 acetylation .......................................................................................... 5 Histone and DNA methylation ........................................................................... 5

H3K9 methylation .......................................................................................... 6

DIFFERENT CHROMATIN STRUCTURES ................................................. 6 Euchromatin ....................................................................................................... 6 Heterochromatin ................................................................................................ 6 GREEN, BLUE, BLACK, RED, YELLOW chromatin ................................................. 8 Position-effect variegation ................................................................................. 9

HP1a (Su(var)2-5) ............................................................................... 10 Repressive or activating function of HP1a? ................................................. 11 Isoforms of HP1 ............................................................................................ 12

MEDIATION OF H3K9 METHYLATION MARKS ...................................... 12 G9a ................................................................................................................... 12 Su(var)3-9 ......................................................................................................... 13 SETDB1 ............................................................................................................. 14

HETEROCHROMATIN FORMATION ...................................................... 14 Heterochromatin formation and RNA interference ......................................... 15 Transposons ..................................................................................................... 16

WHY USE DROSOPHILA AS A MODEL ORGANISM? .............................. 17 General information about the fruit fly ........................................................... 17 Specific advantages of fruit fly in epigenetics .................................................. 19

Polytene chromosomes................................................................................ 19

Table of CONTENTS

ii

Two chromosome-wide regulatory systems ................................................. 19

DOSAGE COMPENSATION .................................................................. 19 General ............................................................................................................. 19 In mammals ...................................................................................................... 20 Up-regulation of mammalian X-chromosome .................................................. 21 In Drosophila ..................................................................................................... 22

MSL1 ............................................................................................................. 22 MSL2 ............................................................................................................. 23 MSL3 ............................................................................................................. 23 MLE ............................................................................................................... 23 MOF .............................................................................................................. 24 roX1 and roX2 ............................................................................................... 24 High affinity sites and spreading of DCC ....................................................... 25 Targeting mechanisms of the MSL complex ................................................. 25 Mechanism behind the X-chromosome up-regulation ................................. 26

Painting of Fourth (POF) ..................................................................... 27 Chromosome 4.............................................................................................. 28 Haplo-4

th lethality and POF ........................................................................... 29

Balanced regulation of chromosome 4 genes by POF and HP1a .................. 29 Evolutionary links between POF and the MSL complex ................................ 29

Do other compensating systems exist? ............................................................ 30

ANEUPLOIDY ..................................................................................... 31 Cancer, developmental diseases and aneuploidy ............................................. 31 Aneuploidy and evolution ................................................................................. 32 Aneuploidy in Drosophila .................................................................................. 33 Are there mechanisms for aneuploidy compensation? .................................... 33 Buffering ........................................................................................................... 33

Feedback regulation ..................................................................................... 36 Feedforward regulation ................................................................................ 36 Inverse dosage effect .................................................................................... 37

Challenges with genome-wide expression analysis .......................................... 37 Reference points ........................................................................................... 37 Skewness....................................................................................................... 37 Limitations in the arrays ............................................................................... 38

AIMS ............................................................................................ 39

RESULTS AND DISCUSSION ........................................................... 40

PAPER I AND II ................................................................................... 40

Table of CONTENTS

iii

General buffering levels ................................................................................... 40 Buffering of specific gene groups ..................................................................... 40

UEGs and NUEGs .......................................................................................... 40 Gene length and wildtype expression level affects buffering ...................... 41

Buffering mechanisms ...................................................................................... 41 POF compensates chromosome 4 ................................................................ 42 Buffering induces proteolysis ....................................................................... 43

Future perspectives .......................................................................................... 43 Conclusions ...................................................................................................... 44

PAPER III ............................................................................................ 45 HP1a has opposing functions on chromosome 4 and in pericentromeric

regions .............................................................................................................. 45 HP1a has different functions at the promoter and at the gene body .......... 45

SETDB1 and Su(var)3-9 are complementary to each other ............................. 46 HP1a displays a stronger repression of long genes .......................................... 47 HP1a effect in the pericentromeric regions depends on location ................... 47 Concluding remarks ......................................................................................... 48 Conclusions ...................................................................................................... 49

PAPER IV ............................................................................................ 50 POF targets roX proximal sites ......................................................................... 50 Connection with the MSL complex .................................................................. 50 Parts of PoX2 functions as POF high affinity target ......................................... 51 HP1a correlates with POF in the PoX sites ....................................................... 51 Conclusions ...................................................................................................... 52

FINAL CONCLUDING REMARKS ........................................................... 53

ACKNOWLEDGEMENTS ................................................................ 54

REFERENCES ................................................................................ 58

PAPER I-IV

LIST OF PUBLICATIONS

iv

LIST OF PUBLICATIONS

This thesis is based on the following papers, which in the text will be referred

to by their Roman numerals (I-IV).

I Per Stenberg, Lina E Lundberg, Anna-Mia Johansson, Patrik

Rydén, Malin J Svensson and Jan Larsson (2009). Buffering of

segmental and chromosomal aneuploidies in Drosophila

melanogaster, PLoS Genet 5:e1000465

II Lina E Lundberg, Margarida L A Figueiredo, Per Stenberg and Jan

Larsson (2012). Buffering and proteolysis are induced by segmental

monosomy in Drosophila melanogaster. Nucleic Acids Res 40:

5926-5937

III Lina E Lundberg, Per Stenberg and Jan Larsson (2013). HP1a,

Su(var)3-9, SETDB1 and POF stimulate or repress gene expression

depending on genomic position, gene length and expression pattern in

Drosophila melanogaster. Nucleic Acids Res doi:

10.1093/nar/gkt158

IV Lina E Lundberg, Maria Kim, Anna-Mia Johansson, Marie-Line

Faucillion, Rafael Josupeit and Jan Larsson (2013). Targeting of

Painting of fourth to roX1 and roX2 proximal sites links dosage

compensation to heterochromatin in Drosophila melanogaster.

Submitted manuscript

Paper I-III are reproduced with permission from the publishers.

The following paper is not included in this thesis;

Filip Crona, Olle Dahlberg, Lina E Lundberg, Jan Larsson and Mattias

Mannervik (2013). Gene regulation by the lysine demethylase KDM4A in

Drosophila. Dev Biol 373:453-463

TERMINOLOGY AND ABBREVATIONS

v

TERMINOLOGY AND ABBREVATIONS

Autosome All chromosomes which are not sex-chromosomes

CD Chromo domain

Centromere Part of the chromosome that links sister chromatids during

metaphase

Chromatin The DNA-protein structure, which all DNA in the nucleus

is present in

CSD Chromo-shadow domain

DCC Dosage Compensation Complex

dsRNA Double-stranded RNA, two RNA molecules that base-pair

with each other

Ectopic Occurring in an abnormal position

Exon The segments of a gene that will be present in the mRNA

when it is transcribed

Gene body The entire gene from the transcription start site to the end

of the transcript

Haplo-4th One copy of the 4th chromosome

Heterozygous Two different alleles for a single trait

Histone Small DNA binding proteins that forms the complex which

DNA is wrapped around in nucleosomes

HKMT Histone lysine methyltransferase

Homozygous Identical alleles of a single trait

HP1a Heterochromatin protein 1 a

H3K9me1, me2, me3 Histone H3 mono-, di-, trimethylated at lysine 9

(associated with inactive genes)

H3K36me3 Histone H3 tri-methylated at lysine 36 (associated with

active genes)

H4K16ac Histone H4 acetylated at lysine 16 (associated with active

genes)

Intron The segments in-between exons in a gene, they are not part

of the mRNA

Kb Kilo base pairs

Mb Mega bases

Mitosis When a cell separates all replicated chromosomes into two

identical groups, before dividing into two daughter cells

Meiosis Special type of cell division that forms eggs and sperm

cells. Four haploid cells are produced from one diploid cell

MOF Males absent on the first

Monosomic A chromosome region present in one copy

MRE MSL recognition element

mRNA messenger RNA, carrier of information from DNA

TERMINOLOGY AND ABBREVATIONS

vi

MSL Male specific lethal, the complex which mediates dosage

compensation of the Drosophila male X-chromosome

nm Nano meter

Nucleotide The building blocks of DNA: A, T, C, G

Nucleosome Basic DNA packaging unit consisting of a histone-DNA

complex

NUEG Non- Ubiquitously Expressed Gene, genes required for

tissue specific functions, not expressed in all tissues

Orthologue Genes in different species that evolved from a common

ancestral gene by speciation

Paralogue Related genes which have occurred due to duplication

within a species

Pericentromeric regions Heterochromatic regions near the centromere

Promoter A region of DNA that initiates transcription from the

nearby located gene

RNAi RNA interference

RNA polymerase II An enzyme that produces RNA from a gene

roX RNA on X, non-coding RNA which is part of the MSL

complex

S-phase The phase of the cell-cycle in which DNA is replicated,

occurs before mitosis

SXL Sex lethal (prevents MSL2 from forming in females)

Su(var) Suppressor of variegation

TE Transposable elements

Transcriptional elongation The process when RNA polymerase II reads a gene and

synthesizes RNA

Transgene A gene or genetic material that researchers have inserted in

a genome, or into another species

Trisomic A chromosome region present in three copies

UEG Ubiquitously Expressed Gene, genes required for

maintenance of basic cellular functions (housekeeping)

Xi Inactivated X chromosome, found in mammals

3’ end The end of a gene, is transcribed last

5’ end The beginning of a gene, is transcribed first

ABSTRACT

vii

ABSTRACT

Stimulation or repression of gene expression by genome-wide regulatory

mechanisms is an important epigenetic regulatory function which can act to

efficiently regulate larger regions or specific groups of genes, for example by

compensating for loss or gain of chromosome copy numbers. In Drosophila

melanogaster there are two known chromosome-wide regulatory systems;

the MSL complex, which mediates dosage compensation of the single male

X-chromosome and POF, which stimulates expression from the

heterochromatic 4th chromosome. POF also interacts with the

heterochromatin inducing protein HP1a, which represses expression from

the 4th chromosome but which also has been assigned stimulatory functions.

In addition to these two, there is another more elusive and less well-

characterized genome-wide mechanism called buffering, which can act to

balance transcriptional output of aneuploidy regions of the genome (i.e. copy

number variation).

In my thesis, I describe the presence of a novel physical link between dosage

compensation and heterochromatin; mediate by two female-specific POF

binding sites, proximal to roX1 and roX2 on the X chromosome (the two

non-coding RNAs in the MSL complex). These sites can also provide clues to

the mechanisms behind targeting of chromosome-specific proteins.

Furthermore, to clarify the conflicting reports about the function of HP1a, I

have suggested a mechanism in which HP1a has adopted its function to

different genomic locations and gene types. Different binding mechanisms to

the promoter vs. the exon of genes allows HP1a to adopt opposite functions;

at the promoter, HP1a binding opens up the chromatin structure and

stimulates gene expression, whereas the binding to exons condense the

chromatin and thus, represses expression. This also causes long genes to be

more bound and repressed by HP1a. Moreover, I show that buffering of

monosomic regions is a weak but significant response to loss of

chromosomal copy numbers, and that this is mediated via a general

mechanism which mainly acts on differentially expressed genes, where the

effect becomes stronger for long genes. I also show that POF is the factor

which compensates for copy number loss of chromosome 4.

SVENSK SAMMANFATTNING

viii

SVENSK SAMMANFATTNING

Alla celler i kroppen innehåller all vår arvsmassa, våra gener, i form av DNA

och funktionen för varje enskild cell styrs av vilka gener som är aktiva

(uttryckta) i just den cellen. Det är därför extremt viktigt att regleringen av

hur gener används fungerar som den ska. Denna reglering sker ofta på en

enskild gen-nivå, men förekommer också på en mer generell nivå på grupper

av gener eller på hela kromosomer. Fördelen med en generell genreglering är

att cellen på ett effektivt och synkroniserat sätt kan reglera grupper av gener

som är kopplade till liknande funktioner eller till samma region, till exempel

genom att kompensera uttrycket om delar av en kromosom av någon

anledning tappar eller får extra kopior, ett tillstånd som kallas aneuploidi.

Detta är vanligt förekommande i naturen och det är till och med troligt att

varje människa bär på hundratals små aneuploida regioner. Gravare

aneuploidi är starkt förknippat med tumörer och utvecklingsstörning, t.ex.

Downs syndrom är orsakat av en kopia för mycket av kromosom 21. Det

finns mekanismer som verkar i aneuploida regioner för att dämpa effekterna

av felaktig gen-dos, så kallad buffring, men hur de fungerar är fortfarande

mycket oklart. Jag visar i min avhandling att den buffring som motverkar

effekterna av en halverad gen-dos troligtvis är en generell mekanism som

känner igen regioner med felaktiga kopienummer, och som dessutom har

starkast effekt på långa gener. Gener som är viktiga för vävnadsspecifika

funktioner, och därmed bara aktiva i vissa celler, verkar också ha lättare för

att bli buffrade och bör alltså klara en halverad gen-dos bättre än gener som

är konstant aktiva och involverade i livsuppehållande processer i alla celler.

Utöver denna ännu ganska oklara buffringsmekanism finns det två mer

väldefinierade, kromosom-specifika system i bananflugan som stimulerar

genuttryck: proteinkomplexet MSL, som doskompenserar hanarnas enda X-

kromosom så att den får dubbelt så högt uttryck och blir likvärdig med

honornas två X-kromosomer, samt proteinet POF, som specifikt binder

kromosom 4 och stimulerar genuttrycket. POF är också starkt kopplat till

HP1a, ett proteins som är mycket viktigt för att cellen ska kunna bilda

heterokromatin, en tätt packad DNA struktur som tystar ner de flesta gener.

Jag visar att POF kan binda till två speciella ställen på honornas X-

kromosom, vilket dels kan vara mycket användbart för att förstå

mekanismen bakom hur POF känner igen kromosom 4, men som också

tyder på att POF har kopplingar till doskompensering. Jag visar också att

HP1a kan ha motsatta effekter på genreglering; om HP1a binder till en

promoter (en DNA-sekvens bredvid genen som hjälper till att reglera

uttrycket) uppstår en löst packad DNA struktur som leder till ökat

genuttryck, medan HP1a bindning till själva genen inhiberar genuttryck.

INTRODUCTION

1

INTRODUCTION

The processes behind gene expression are of course extremely complex, with

many different proteins involved in recognizing, binding to, reading, and

ultimately translating the messages within the genes, the DNA code, into

fully functional proteins that are required in different processes of the cell.

The cell nucleus is literally packed with proteins with the only purpose of

maintaining a balanced and functional flow of information from the genes.

All this would be far too complicated to describe in just one book, so this

thesis will be focused on one small part of the puzzle: genome-wide gene

regulation, which involves how the regulation of gene activation, or

repression, can be maintained on a larger scale. This means not only

regulation of individual genes but rather, the regulation of large genomic

regions, and even entire chromosomes. Most of this works falls under a

branch of genetics called epigenetics.

EPIGENETICS

The properties and characteristics of all life forms are said to depend on the

genetic factor, the DNA code containing all the heritable information, as well

as on the environmental factor which shapes us during life. Which one is the

actually determining factor for many of our traits is very often a subject of

dispute, but it has grown more and more evident that it is likely a

combination of both, and that they are linked by an intermediate factor: the

epigenetic factor.

Examples to illustrate the impact of epigenetics: compare one of your brain

cells with one of your skin cells or muscle cells; they contain the exact same

DNA, and yet they have such different features, or compare two adult

identical twins; they have identical DNA but will most likely differ in small

physical characteristics. This is caused by epigenetics.

Epigenetic literally means “above genetic” and is defined as changes and

patterns in gene expression, which can be inherited through mitosis and

sometimes also meiosis, but which are not caused by any changes in the

actual DNA sequence.

To explain in other words: The DNA code is (ideally) a fixed and non-

changing sequence of letters which, when put in different combinations (a

bit like the binary system in computers), can be interpreted into information

INTRODUCTION

2

that the cell uses to assemble proteins. But this does not mean that all parts

of this code are always read at all times (much like in a computer).

Depending on the cell and the function it has, different genes will be read,

and at different times.

So how can your body (or the bodies of identical twins), have emerged from

just one single cell and turned into all these different cell types that have

adopted so different fates. And an equally important question is: how do they

remember their fate throughout endless rounds of cell division?

Cell differentiation

The initiation of cell differentiation (i.e. establishing the fate of a cell) is

usually a complex orchestra of different signal molecules that permeate the

embryo during early development. In Drosophila (i.e. fruit fly), this process

is well studied and it all begins with a few number of signal molecules that

are deposited in opposite ends of the egg (oocyte) by the mother. These will

define the polarity of the embryo (i.e. determine the anterior- posterior axis)

and diffuse throughout the embryo, creating a gradient of the signal

molecules which will result in the activation or inhibition of various different

genes. These genes have evolved to respond differently (i.e. be active or

silent) to different concentrations of signal molecules, hence the position a

cell has within the embryo will determine its fate, “simply” by the different

gradients of the signal molecules that reaches the cell. This cell will then in

turn activate new genes which will submit new signal molecule gradients,

and thus the complex process of cell differentiation begins.

However, once the fate of a cell is established, it needs to maintain this

identity even in the absence of the initial signal. This is actually where

epigenetics comes into the picture; epigenetics is the memory by which

expression statuses assigned to various genes and/or genomic regions are

maintained through cell division. Essentially, this memory is mediated by

keeping active genes easily accessible to the transcription machinery,

whereas inactive genes are blocked from transcription. The key to this lies in

the organization of the DNA in the nucleus, and the various proteins which

are surrounding the DNA.

CHROMATIN

To really understand the concept of epigenetics, it is important to first

understand the organization of the chromosomes and the genetic material;

INTRODUCTION

3

since the DNA material within each individual cell measures a total of about

2 meters in humans, it is extremely important for the cell to keep the DNA

organized to uphold a smooth and efficient reading of the genetic material,

and a correct replication and distribution of the DNA into two daughter cells

during cell-division. This is done by packaging all DNA into a large,

condensed DNA-protein structure called chromatin. As a first step of

condensation, stretches of DNA are wrapped around a specific protein core

complex, forming a DNA-protein complex called a nucleosome (see figure

1A), which is found in all eukaryote genomes [1-3]. The protein core, also

called the histone core, is made up of an octamer of the different histone

proteins H3, H4, H2A and H2B, which are all present in two copies [4].

Around the histone octamer, 146 base pair (bp) of DNA are wrapped forming

1.65 turns [5] and the DNA is then “locked” in position on the histone core by

a linker histone protein 1 (H1) [6]. The nucleosome structure compacts the 2

nm thick DNA about a five- to ten-fold, into an 11 nm thick structure (see

figure 1B). Between each nucleosome is a short segment of linker DNA

which, depending on level of compaction, varies between about 10-80 bp in

length [7,8], giving the DNA a typical “beads on a string” like structure. This

structure is best visible if the linker histone 1 is removed [1-3].

This structure is then further condensed into an about 50-times folded, 30

nm thick fiber-structure [9] (see figure 1B), by either folding into a zigzag

structure or into a solenoid-like structure [1]. In a mitotic chromosome, a

partly unknown mechanism can condense this 30 nm structure even a few

hundred-fold more, to form the final compacted structure which makes the

chromosomes visible as separate units (see figure 1B), or the “X”-like shape

that we associate with a metaphase chromosome in a karyogram (the “X” is

formed when the cells are about to divide and each chromosome has

replicated into two copies, only attached by the centromere).

The histone core, on which the DNA resides, is the actual backbone for

almost all the epigenetic information that can control gene expression. This

is because each histone molecule has 15-38 amino acids protruding in the N-

terminal end, forming a “tail” domain, which is accessible outside of the

compact DNA-histone core [10]. These histone tails function as potential

modification site where a variety of different small tags can be attached by

different enzymes. These tags can affect the biological role of the underlying

DNA by modifying the level of chromatin compaction, often by functioning

as signals, or binding platforms, for specific chromatin-associated

proteins, which can bind to the nucleosome and affect the properties of the

local chromatin structure; either making it more condensed, and thus silent,

INTRODUCTION

4

or making it more loose, which increases transcription activity. One example

of such a protein is HP1a, which will be mentioned later on in this thesis.

Several different types of tags have been found on the different histones, for

example: methylation, acetylation, phosphorylation, ubiquitylation and

sumoylation. Depending on which amino acid-residue they modify and at

which position on the histone tail, they will have different functions. These

modifications are dynamic and rapidly changing, they can be added or

disappear within minutes of stimuli on the cell surface. I will go into a

selection of these modifications that are relevant for this thesis:

Figure 1. DNA compaction. A) The nucleosome core, consisting of an octamer of two

copies each of four histone molecules. DNA (light blue) is wrapped around the core, and one

copy of the linker histone H1 secures the DNA binding. Each histone has an N-terminal tail

protruding out of the core, which can be labeled by different tags (for example a methyl

group). B) The different compaction levels of chromatin (i.e. the DNA-histone complex that

constitutes the chromosomes), from the 2 nm thick naked DNA helix to the 1400 nm thick

highly condensed metaphase chromosome.

A B

INTRODUCTION

5

Histone acetylation

Acetylation is generally associated with active chromatin and of the known

modifications, acetylation has the highest potential of unfolding chromatin,

since it can neutralize positive charges on the target residues and thus make

the tail of the histone less prone to interact with the negatively charged DNA.

This allows easier access of the transcriptional machinery to increase

transcription [11].

H4K16 acetylation

An example of acetylation is the covalent addition of an acetyl group onto the

16th amino acid, a lysine (K), on the tail of histone 4 (H4K16ac), which is

highly associated with the action of the dosage compensation system in male

Drosophila (this will be discussed in detail later). The acetylation of lysine 16

on histone 4 [12] has also been shown to inhibit the formation of the 30

nanometer fiber and higher-order chromatin structures [13] [14].

Histone and DNA methylation

Methyl groups can be covalently added to the histone tails of nucleosomes,

but also directly to the DNA. DNA methylation is a repressive mark highly

associated with epigenetic inheritance in mammals, however, the role of

DNA methylation in Drosophila is more elusive. Some traces of this has been

found in early stages of embryonic development [15], but in general, it is

believed not to have any functional significance in Drosophila and therefore;

I will focus on the methylation of histones. This modification can have both

positive and negative effects on the level of transcriptional activity and this

depends on the position of the target residue in the tail. Furthermore, this

modification adds even extra complexity because each histone tail binding

site can have between one and three methyl groups attached: Lysines (K) can

be mono- (me1), di- (me2) or tri (me3)- methylated and arginines (R) can be

mono- or di- methylated [16].

There are so far six well characterized methylation marks, three of which are

in general correlated to active transcription: H3K4 [17], H3K36 [18] and

H3K79 [19]. The other three are associated with transcriptional repression:

H3K9, H3K27 and H4K20 [20]. Most important for this thesis is the H3K9

methylation:

INTRODUCTION

6

H3K9 methylation

Methylation of lysine 9 on the tail of histone 3 results in repression of gene

expression and is highly associated with heterochromatin. In Drosophila, it

is primarily found on chromosome 4, in the centromeric and

pericentromeric regions. It is mediated by the proteins Su(var)3-9 and

SETDB1 and is essential for the binding of a protein named HP1a, these

will all be described in detail later.

DIFFERENT CHROMATIN STRUCTURES

Chromatin can be classified into different types depending on the different

levels of chromatin compaction, level of gene activity, and associated histone

modifications. Traditionally two different main types have been defined:

euchromatin and heterochromatin.

Euchromatin

The active parts of the genome essentially consists of euchromatin, a

structure in which the DNA is loosely packed around the histone cores (see

figure 2A) and this allows access of the transcriptional machinery to the

genes. Euchromatin is loosely packed throughout most part of the cell cycle

(interphase), and only becomes condensed during the relatively short mitotic

phase of the cell cycle when the duplicated DNA needs to be efficiently

distributed between the two daughter cells. Known histone modifications

associated with active chromatin are: H3K4 methylation, H3K9 acetylation

and H3S10 phosphorylation [21-23].

Heterochromatin

Heterochromatin is considered to be the “silent” chromatin, associated with

very little gene activity. It is a gene-poor, condensed structure (see figure

2A), which generally lacks mitotic recombination, is late replicating, and

remains condensed throughout the entire cell cycle. It is usually associated

with low levels of acetylation and high levels of some methylated sites such

as H3K27, H4K20 and most importantly, methylation of H3K9 [21-23].

Interestingly, the level of heterochromatinization can be affected by

environmental factors, for example temperature changes in Drosophila

during development; a decrease in temperature from 25˚C to 18˚C results in

more heterochromatic silencing of gene expression, whereas an increase in

INTRODUCTION

7

temperature to 29˚C results in less heterochromatic silencing [24]. Other

factors that affect the rate of development also show similar effects. The

highly heterochromatic Y-chromosome is known to affect the level of

heterochromatinization; flies carrying an extra copy of chromosome Y show

reduction in heterochromatic silencing in other parts of the genome. On the

other hand; male flies lacking the Y-chromosome (X0) display enhanced

heterochromatic silencing [25,26].

About one-third of the Drosophila genome consists of heterochromatin and

regions, which are located close to the centromeres, most parts of

chromosome 4, and the telomere regions [27] (see figure 3A). The

localization of heterochromatin might be important for protection of DNA

during replication and to separate sister chromatids in mitosis. The

pericentromeric regions and chromosome 4 are predominantly associated

with H3K9me2, whereas the centromeric regions are most enriched in

H3K9me3 [22,28]. The Y-chromosome, which corresponds to about 20-30%

of the male genome (~40 Mb) in size, is also considered to be

B

A

Figure 2. Heterochromatin vs. euchromatin. A) Schematic illustration of the

difference in compaction between heterochromatin and euchromatin, and the possible role

HP1a has in compacting chromatin into heterochromatin. HP1a binds to methyl groups on

lysine 9 on the N-terminal tail of histone H3 (H3K9me) and forms an HP1a dimer that links

together two adjacent nucleosomes. B) Schematic illustration of the protein-domains of

HP1a.

INTRODUCTION

8

heterochromatic and mostly contains repetitive elements and very few genes

[29]. In addition, small regions of heterochromatin are found dispersed in

the euchromatic parts of the genome.

Since the amount of genomic heterochromatinization is variable and can be

affected by factors such as temperature, it is possible that euchromatin and

heterochromatin are highly dynamic states that are sometimes more

transient, and sometimes more fixed, and that large heterochromatic regions

have the role of titrating the amount of heterochromatin associated proteins.

If the amount of heterochromatin proteins becomes too large, threatening

the expression of the active genome, it can be redirected to heterochromatic

regions, such as the Y-chromosome. On the opposite, if the Y-chromosome is

missing, the amount of heterochromatin proteins in the genome becomes

abundant, and can repress other regions.

GREEN, BLUE, BLACK, RED, YELLOW chromatin

The traditional view on chromatin is that heterochromatin and euchromatin

are the two main types of structure, but it has become more and more

evident that there are large variations also within these groups, and therefore

a more specific classification of chromatin has been proposed [30], in which

five principal chromatin types, defined by unique combinations of associated

chromatin binding proteins, are specified in Drosophila:

GREEN chromatin corresponds to “classical” heterochromatin and is

found primarily in pericentromeric regions and on chromosome 4. It is

defined by the presence of Su(var)3-9, HP1a (described later on in this

thesis), as well as a couple of HP1a interacting proteins: LHR and HP6. In

addition, GREEN chromatin is also highly and specifically enriched in

H3K9me2.

BLUE chromatin corresponds to PcG chromatin since it is associated with

the Polycomb-group (PcG) proteins PC, E(Z), PCL and SCE (these are

proteins important for maintaining silencing of specific genes during

Drosophila embryonic development). In addition, it is highly enriched with

the repressive histone modification H3K27me3, which is mediated by E(Z)

and recognized by PC. The Hox gene cluster (a well-known PcG target loci) is

found within BLUE chromatin.

BLACK chromatin is the most abundant chromatin type, covering 48% of

the probed genome of Drosophila. It is generally gene poor, and the enclosed

INTRODUCTION

9

genes (4162 genes) are either silent or expressed at very low levels. In fact, a

majority of the silent genes in the genome are found within the BLACK

chromatin definition. It lacks the active histone marks H3K4me2 and

H3K27me3 but is marked by a number of chromatin associated proteins, like

histone H1. BLACK chromatin appears to have an active role in

transcriptional silencing, since it has a higher ability to repress inserted

reporter genes than the rest of the genome.

RED chromatin and YELLOW chromatin both constitute the active

chromatin (euchromatin) and they share several marks for active chromatin.

Both are replicated early in S phase of the cell cycle, but characteristic for

RED chromatin is that it tends to be replicated even earlier in S-phase and is

strongly enriched in origin of replication complex (ORC). Apart from

replication timing, the YELLOW chromatin differs from RED chromatin by

being enriched in MRG15, a chromo domain-containing protein that has

been suggested to bind to H3K36me2 and me3, [31] and YELLOW

chromatin in general also contains more H3K36me3 than RED chromatin.

Genes that are ubiquitously expressed over many embryonic stages and

tissue types, such as ribosomal- and DNA repair genes and genes involved in

nucleic acid metabolic processes, are mostly found within YELLOW

chromatin whereas genes involve in more specific processes such as defense

responses, signal transduction etc. are more located in RED chromatin.

These five different types of chromatin are distributed in a mosaic around

the entire genome, with accumulation of repressive chromatin around the

centromeres. Although this classification is rough, it provides a more specific

definition of different chromatin types, which often differs extensively in

characteristics. A similar classification has been done by the modENCODE

project, in which they defined nine chromatin states [32]. Chromatin state 7

or the GREEN chromatin corresponds to the heterochromatin type that will

be in focus in this thesis.

Position-effect variegation

A chromatin connected mechanism, which has been particularly well studied

in Drosophila, is position-effect variegation, one of the first mutations that

was discovered in Drosophila was white, which changes the eye color of the

fly from the normal red into white. Muller experimented with X-ray as a

mutagen to induce mutations and discovered an unusual and interesting

phenotype; a fly with red and white patched eyes, i.e. variegated eyes. This

was highly intriguing since it meant that the white gene was indeed affected

INTRODUCTION

10

by the mutagenesis, but not destroyed. Had it been destroyed, every facet of

the eye would have been white. Moreover, this variegated phenotype could

be restored back into wildtype red eyed in some of the offspring of variegated

parents, by again using radiation. This meant that it was no ordinary

mutation, and to better understand this, they examined the chromosomes of

these variegated flies and found that a chromosome breakage and

subsequent inversion of a chromosomal segment had translocated (i.e.

moved) the white gene into close vicinity of heterochromatin. The gene was

thus still intact and could in principal be expressed, but the new

heterochromatic silencing surrounding could actually spread into the white

gene, causing varying degree of repression of the red eye pigmentation which

results in red and white patched eyes. This phenomenon was named

Position-effect variegation (PEV) (reviewed by [33]), since this

phenotype was a result of a change in position of the white gene. In

Drosophila, practically every gene examined has shown PEV when

translocated into the vicinity of heterochromatin. PEV of the white gene in

Drosophila has been used as a tool to identify genes involved in

heterochromatin formation; by combining PEV flies with various other

mutations, scientists could search for those mutations that result in an

increase of silencing of the reporter gene, termed Enhancer of variegation,

E(var) or in a decrease of silencing, termed Suppressor of variegation,

Su(var). Using this method, so far about 30 different genes involved in

modifying PEV have been isolated and characterized [33].

Two of these genes have particular importance for the formation of

heterochromatin and the silencing of genomic regions: Su(var)2-5, encoding

the protein HP1a and Su(var)3-9, encoding the protein Su(var)3-9, which

mediates a methylation mark required for HP1a targeting. I will describe

both in more detail.

HP1A (SU(VAR)2-5)

HP1a (Heterochromatin Protein 1a) is the most important factor involved in

establishing and maintaining heterochromatin by binding directly to and

affecting the compaction level of the chromatin. A complete removal of HP1a

using a homozygous mutant results in loss of heterochromatin and hence,

lethality [34]. HP1a essentially functions by binding to H3K9me2 and me3

marks and initiates a compaction of the nucleosomes. The protein is made

up of two functionally important parts: the N-terminal chromo domain (CD),

which mediates the interaction with H3K9me2 and me3 [35,36] [37], and

the C-terminal chromo-shadow domain (CSD), which can interact with the

INTRODUCTION

11

CSD of another HP1a molecule to create a HP1a dimer [38,39]. The two

functional domains are connected through a hinge domain, which gives the

protein flexibility and potentially aids in the recognition of H3K9

methylation [40] (see figure 2B).

HP1a binding sites in the Drosophila genome are highly associated with

GREEN chromatin [30] and are found primarily in the pericentromeric

region, on the 4th chromosome, and dispersed in a number of smaller

euchromatic sites, of which cytological region 31 on chromosome arm 2L is

the most distinct one [27,41-45]. On chromosome 4, HP1a is bound to the

gene body (the part of the gene that is transcribed) and the promoters (the

region which recruits transcription proteins and initiates transcription of the

gene) of active genes [30,43-47], whereas in the pericentromeric regions,

HP1a appears to bind in a more general manner and it spreads over tens of

kb rather than to individual genes [45,47].

Interestingly, HP1a binding to the gene body is dependent on H3K9me2 and

me3 whereas the binding peak at the promoters of the genes is methylation-

independent [45], this could indicate that HP1a binds to chromatin by two

different binding mechanisms.

Repressive or activating function of HP1a?

HP1a is, as already mentioned, essential for formation and maintenance of

the highly repressive heterochromatin, and has thus been assumed to have a

repressive function. This view was also supported by an RNAi mediated

knock-down of HP1a, which resulted in an increased expression of genes on

the 4th chromosome [48,49].

But more recently, several conflicting studies have emerged, providing

different evidence for an activating function of HP1a; knock-down

experiments using both RNAi [50,51], and mutation have discovered several

euchromatic genes that, as a result of the HP1a depletion, are down-

regulated in expression [42,52-54]. This is the opposite of the effect you

would expect given HP1a’s inhibiting properties. In addition, genes such as

light, rolled, RpL15 and Dbp80, which are all located within the

heterochromatic pericentromeric regions, have shown to actually depend on

HP1a and the heterochromatic surrounding to maintain proper expression

[55-60]. Furthermore, the existence of euchromatic binding sites of HP1a

also indicates an association with active gene transcription [41] [47,52],

although here, the binding seems mostly localized to the gene body rather

INTRODUCTION

12

than the promoter [52] and is mostly independent on Su(var)3-9 [42]. Other

sites with high gene activity, such as developmentally regulated genes and

heat-shock induced chromosomal puffs, have also been associated with HP1a

binding [52].

All these evidence has led to an unresolved debate whether HP1a has a

repressing or a stimulating function on gene expression, and this is an issue

explored and discussed in paper III in this thesis.

Isoforms of HP1

D. melanogaster encodes no less than 5 paralogues of HP1: HP1a, HP1b,

HP1c, HP1d/Rhino and HP1e. They are structurally similar, containing the

conserved chromo domain and chromo-shadow domain that are separated

by a variable hinge domain. Yet they appear to have different functions;

HP1a is the most studied isoform and usually the one that is referred to in

general studies of HP1 function. As described above, HP1a localizes mainly

with H3K9 methylation and the GREEN heterochromatic regions. HP1b is

found in both euchromatic and heterochromatic regions whereas HP1c is

primarily localized to euchromatic regions [61], and co-localizes with

H3K4me and RNA polymerase II (both marks for active gene transcription)

[62]. HP1a, HP1b and HP1c are all ubiquitously expressed whereas

HP1d/Rhino and HP1e are mainly functioning in the germline of males and

females [63].

MEDIATION OF H3K9 METHYLATION MARKS

As mentioned above, the binding of HP1a to the chromatin and the gene

bodies of active genes requires the presence of the histone modification

H3K9me2 and/or me3, i.e. the attachment of two or three methyl-groups to

lysine (K) 9 on the tail of histone H3. The exception is HP1a binding to

promoter regions, which is independent on H3K9 methylation. The proteins

that mediates these methylations are known as histone lysine methyl

transferases (HKMTs) and in Drosophila there are three known HKMTs;

G9a, Su(var)3-9 and SETDB1.

G9a

The Drosophila dG9a protein appears to be a functional orthologue of the

mammalian G9a, which mediates mono- and dimethylation in euchromatic

regions [64], however, in Drosophila G9a does not appear to affect the H3K9

INTRODUCTION

13

methylation patterns and is also not important for viability [65]. A role in

germ cell formation has been proposed [66], but in general, the function of

G9a in Drosophila remains elusive [45,65,67].

Su(var)3-9

Su(var)3-9 controls H3K9me2 and me3 [68] primarily in the centromeric

and pericentromeric regions, and is therefore important for gene silencing

through formation of heterochromatin [69-71]. Notably, in the centromeric

regions, Su(var)3-9 only seem to control the me3, but not the me2 [20,70].

It actually is a rather funny coincidence that this protein happens to

methylate H3K9, since the name 3-9 was given because the gene is situated

on chromosome 3, and 9 was just a serial number applied during the

screening experiment that identified it. The actual function was discovered

much later.

Su(var)3-9 mutants have depleted levels of HP1a and H3K9me2 in the

pericentromeric regions and, in contrast to HP1a mutants, are generally

viable, fertile and unaffected in germ-cell development [71]. A mutation that

gives the protein a hyperactive function results both in stronger PEV (i.e.

heterochromatic silencing of reporter genes), higher levels of H3K9me2 and

me3 at the chromocenter, and also ectopic H3K9 methylation in several

euchromatic sites. Su(var)3-9 contains a SET domain, which mediates the

methylations, and a chromo domain, commonly associated with chromatin

remodeling proteins [70,71]. Both the SET domain and the chromo domain

are required for the binding to heterochromatic sites [70].

The presence of Su(var)3-9 and its methylation of H3K9 in heterochromatic

regions is a prerequisite for HP1a binding to these regions, but in fact, these

two proteins are interdependent on each other; HP1a is essential for

restricted Su(var)3-9 binding to heterochromatin and the two proteins can

bind to each other through interactions between the chromo domain of

Su(var)3-9 and the chromo-shadow domain of HP1a [70]. Furthermore, in a

HP1a mutant background, the H3K9 methylation patterns are increased,

particularly in euchromatic sites [20].

In addition to the pericentromeric regions, Su(var)3-9 binding has been

detected on chromosome 4, but the H3K9 methylation pattern of the 4th

chromosome is unaffected in a Su(var)3-9 mutant, so the function of

Su(var)3-9 here is still unknown [20,45]. Furthermore, in contrast to

Su(var)3-9 binding to pericentromeric regions, it appears as if Su(var)3-9

INTRODUCTION

14

does not require the SET domain and the chromo domain to bind to

chromosome 4 [70].

SETDB1

The other HKMTs that mediates H3K9me2 and me3 in Drosophila is

SETDB1 [45,54,72-74], which is encoded by the gene Setdb1 (also named

eggless or egg) [64]. Su(var)3-9 and SETDB1 appear in principle to be

complementary and are responsible for H3K9 methylation in different

regions. So whereas Su(var)3-9 mediates H3K9 methylation in the

pericentromeric regions, SETDB1 is mainly responsible for mediating H3K9

methylation on chromosome 4 [45,72], although it has also been suggested

to mediate H3K9 methylation at some euchromatic sites [72]. In a Setdb1

mutant, H3K9 methylation and HP1a binding is impaired on chromosome 4,

but not in pericentromeric regions [73].

More specifically, SETDB1 methylates the gene body of active genes on

chromosome 4 [45], and in a Setdb1 mutant, HP1a binding to the gene body

of chromosome 4 active genes is impaired, while HP1a binding to the

promoters is still unaffected [45].

SETDB1 contains a methyl-CpG-binding domain (MDB), which seems able

to recruit deacetylase (HDAC) complexes [75] and a PreSET/SET domain,

which mediates the methylation function [64,73,74]. SETDB1 is essential for

female fertility [74,76,77] by mediating H3K9me3 in germ cells and somatic

cells of the germarium, and early stages of egg chamber require SETDB1 for

proper formation [74]. There are some results that indicate that in the germ

cells, SETDB1 is the only active HKMTs and is therefore responsible for

mediating the H3K9 methylation in the pericentromeric regions (instead of

Su(var)3-9), and then in a later stage of the oogenesis, this function is

transferred to Su(var)3-9 [76].

In addition to the pericentromeric regions and chromosome 4, H3K9

methylation is also detected at telomeres and at some euchromatic sites, but

neither Su(var)3-9 nor SETDB1 appear to be involved in establishing these

marks, indicating the presence of additional unknown HKMTs [20,70].

HETEROCHROMATIN FORMATION

Once H3K9me2 and me3 has been established by either SETDB1 or

Su(var)3-9, they function as docking sites that recruit HP1a [35,36,78,79]

INTRODUCTION

15

through an interaction between the H3K9me2 or me3, and a hydrophobic

pocket in the HP1a N-terminal chromo domain (CD) [35,36,80]. Two

adjacent nucleosome-bound HP1a molecules can interact with each other’s

C-terminal chromo-shadow domain (CSD) to form an HP1a dimer [39],

which causes two HP1a bound nucleosomes to be linked tightly together [38]

(see figure 2A). Depending on regions, Su(var)3-9 or SETDB1 is then

believed to interact with the CSD of the bound HP1 [70] to promote further

methylation of the neighboring nucleosome, forming a new binding site for

HP1a and thus, initiate a spreading mechanism that will cause the chromatin

to become condensed and inactive.

For euchromatin to be heterochromatinized, specific de-methylation, de-

acetylation and de-phosphorylation reactions need to take place within the

euchromatin, and this process seem to begin with the de-acetylation of H3K9

by the enzyme HDAC1 [81]. HDAC1 and Su(var)3-9 have been shown to

associate in vivo, and could thus provide extra complexity to the process of

heterochromatin formation [69].

Heterochromatin formation and RNA interference

In a number of eukaryotes, RNAi is involved in heterochromatin formation.

For example in S. pombe (fission yeast), RNAi is required for spreading of

H3K9 methylation into reporter genes that have been inserted into

heterochromatic regions [82]. RNA interference or RNAi was originally

identified as a mechanism involved in post-transcriptional regulation and

defense against for example retroviral infections. It is also believed to be

involved in post-transcriptional silencing of transposable elements. The

RNAi mechanism regulates the levels of RNA transcript in several steps; first

a group of proteins known as the dicer family targets and cleaves dsRNA into

short segments of 21-30 nucleotides, called small interfering RNAs (siRNA).

These siRNAs are then binding to argonaute proteins to form the RISC (RNA

Induced Silencing) complex. The siRNA will base pair with complementary

transcripts and hence, guide the RISC complex to degrade it.

Recently, an RNAi mediated guidance mechanism for targeting of chromatin

remodeling factors to heterochromatic sites has been proposed in

Drosophila, in which the protein Piwi (part of the argonaute protein family)

associates with, and utilizes various different piRNAs (Piwi-interacting

RNAs) to target piRNA complementary sites within heterochromatic regions,

and then recruiting HP1a and Su(var)3-9 to these sites to induce

heterochromatinization. The involvement of the RNAi machinery and small

INTRODUCTION

16

interfering RNAs may explain the lack of sequence recognition domains in

the chromatin remodeling proteins [83].

Transposons

Transposons or transposable elements (TE) are found in both eukaryotic and

prokaryotic genomes and are highly associated with heterochromatic

regions. They constitute a very special type of DNA sequences, also known as

“selfish” elements or as “jumping” genes [84]. Transposable elements can

have the ability to self-replicate and to change position within the genome,

some of them carry within them genes encoding proteins such as

transposases, which can cut and paste the transposon between genomic sites

(DNA transposons), other TE transcribe RNA which is converted into DNA

by a reverse transcriptase, and then inserted into new genomic sites

(retrotransposons). In active state, these elements can potentially cause

severe damage if the insertion site happens to be within an essential gene.

Fortunately, most TE are silent, either due to disruption of their important

genes, or by silencing mechanisms that have evolved in the host genome as a

response to the deleterious effects of active TE [85]. In humans for example,

RNA interference is important for silencing of transposons; small RNAs

from the transposon sequence are incorporated into the RISC complex,

which then can target and repress all the copies of that particular TE [86].

Chromatin modifications also appear to be involved in the repression of TE;

a big proportion of HP1a and H3K9me3 binding is reported to occur to

transposons and repeated elements, where they co-localize with each other.

Transcription activity marks (such as H3K9ac and RNA Polymerase II) on

the other hand, are more or less absent [44], showing that most transposons

in Drosophila, with few exceptions, are transcriptionally silent and that

HP1a is suggested to be involved in this silencing [87] by binding to

Su(var)3-9 mediated H3K9me3 marks [88].

Over the course of evolution, silenced TE seem to have accumulated within

heterochromatic regions, and about 77% of Drosophila heterochromatin (or

about 30% of the entire Drosophila genome) consists of transposable

elements or repeated regions [89], and organisms with larger genomes, for

example humans or maize consists of as much as 45% and 75% transposons

respectively, in the entire genome [82].

If these elements become active and change positions, they sometimes bring

part of the surrounding genome along for the ride. This of course can be very

INTRODUCTION

17

deleterious to a cell, but also, can be a driving force for evolution, and

Drosophila genetic researchers are often utilizing this feature to create

disruptions of genes or genomic segments for studies of gene function.

WHY USE DROSOPHILA AS A MODEL ORGANISM?

General information about the fruit fly

Drosophila melanogaster, or fruit fly, is a black and yellow fly with bright

red eyes, which is usually found around the fruit section in grocery stores, or

in the kitchen of your average student habitat. With a humble size of only 2.5

mm, it might appear very insignificant, or at most, slightly annoying. But in

fact, the fruit fly is very valuable for many researchers and it has been used

as a genetic model organism for more than 100 years (Thomas Morgan Hunt

began to use them for heredity studies around 1910). It became a popular

model organism for many different reasons; the most obvious is that they are

easy and cheap to cultivate and don’t require much space, and with a

generation time of about 10 days at room temperature, you can study several

generations within a few weeks. Genetically they also have several useful

traits; the complete genome of D. melanogaster was sequenced and first

published in 2000 [90]. They have nearly as many genes as humans, about

15 000 genes (compared to 20 000 in humans) and about 75% of human

disease genes have homologues in the genome of Drosophila [91], yet the

size of its genome is substantially smaller and easier to handle, about 260

Mb (2.6 x 108) (according to Drosophila annotation release 5) divided

between just four pairs of chromosomes (to compare with the human

genome which is about 6 x 109 bp divided between 23 pairs of

chromosomes). The genome consists of one pair of sex chromosomes: XX or

XY, and three pairs of autosomal chromosomes: chromosome 2, 3 and 4, of

which chromosome 2 and 3 are each divided into two chromosome arms, left

and right, separated approximately in the middle of the chromosome by the

centromere (they are more specifically named chromosome 2L, 2R, 3L and

3R) (See figure 3A).

INTRODUCTION

18

The Drosophila males are lacking meiotic recombination, and this facilitates

genetic crosses, and because the fly has been used as a model organism for so

many years, a great deal of genetic tools have been developed which are now

readily available for many different uses. For example, genetic markers such

as genes for eye color, body- and wing features etc., which makes it visibly

possible to keep track of the genetic contents of the fly. “Balancer

chromosomes” also exist, which are chromosomes with multiple inversions

that makes it possible to uphold fly strains containing recessive lethal alleles

of genes of interest, which would otherwise quickly vanish from the strain.

Figure 3. The genome of Drosophila melanogaster A) The genome consists of two

pairs of large autosomal chromosomes, 2 and 3 (blue), both divided into left and right arms

separated by the centromere (black dot), one pair of the small autosomal chromosome, 4

(red), and one pair of sex chromosomes, X (green) and Y. The grey areas represent the major

heterochromatic regions primarily consisting of the telomeres (the tip of the chromosomes),

the whole Y-chromosome, parts of chromosome X, most part of chromosome 4, and the

pericentromeric regions (the grey areas around the centromere of chromosome 2 and 3). B) A

polytene chromosome from a male Drosophila, where the DNA is stained with blue and the

two chromosome-wide regulatory systems MSL and POF are stained in green and red

respectively. The heterochromatic areas, shown in grey in (A), are not endoreplicated and are

therefore not visible in a polytene chromosome.

B

A

INTRODUCTION

19

Specific advantages of fruit fly in epigenetics

However, there are two specific properties that are unique for the fruit fly,

which makes them specifically good as a model organism for my work:

Polytene chromosomes

One specifically useful feature of Drosophila, when it comes to studying

chromatin binding proteins, is their polytene chromosomes. These are

extremely large chromosomes found primarily in the salivary glands of third

instar larvae of Drosophila and other dipteran (two-winged) insects. The

purpose of these giant chromosomes is believed to be to increase expression

of proteins that are needed for pupation. The polytene chromosomes are

formed through repeated rounds of DNA replication, but instead of the

normal division into two cells, the cell grows in size and the sister

chromatids remain synapsed together, forming a chromosome consisting of

multiple copies (in Drosophila sometimes up to 1024 copies) adjacent to

each other (a process known as endoreplication). As a result, a very long and

thick chromosome (see figure 3B) is easily seen in microscopes and it is very

good for immunostaining and in situ hybridization, which allows

visualization of the binding sites of proteins and RNAs.

Two chromosome-wide regulatory systems

The second thing that makes Drosophila good for studies of chromatin

associated and chromosome-wide gene regulation, is that it has two known

chromosome-wide gene regulatory mechanisms; the MSL complex, which

decorates the male X-chromosome and mediates dosage compensation,

and POF, a chromosome 4 specific stimulating protein (see figure 3B). POF

is the first and only autosome-specific gene regulatory protein reported in

any organism to this day and therefore makes Drosophila unique compared

to other model organisms. I will go through these two mechanisms in detail:

DOSAGE COMPENSATION

General

In many organisms, sex is determined by the combination of the sex

chromosomes X and Y. Females have the combination XX and males have

the combination XY. These two chromosomes differ substantially in their

genetic content. The Y-chromosome mostly consists of heterochromatin and

INTRODUCTION

20

repetitive elements and very few genes, in mammals about 80 protein-

coding genes [92], out of which the most important one is the key gene in

male sex determination, SRY [93]. The Drosophila male Y-chromosome is

entirely heterochromatic and contains even fewer genes, between 12 to about

20 genes, of which several have suggested male related functions, but unlike

the mammalian sex determination, the Drosophila Y-chromosome is not the

key factor for male determination, but rather for fertility.

The Y-chromosome originally started out as a homologue to the X-

chromosome, the proto-Y, but over the course of evolution, the proto-Y-

chromosome has degenerated in the heterochromatic Y-chromosome,

leaving only a few genes and a small region of homology so that X and Y can

pair during cell division [29,94]. This means that an imbalance in X-linked

genes has occurred between the sexes (♂X: ♀XX) which need to be

compensated for. This is achieved by a dosage compensating mechanism,

which has the purpose of equalizing the gene expression from the X-

chromosome between males (XY) and females (XX) (see figure 4). Different

organisms have evolved different dosage compensating mechanism [95], but

I will focus on the mechanisms in mammals and in Drosophila.

In mammals

Mammals have evolved a mechanism in which one of the two female X-

chromosomes, Xi, in each cell is randomly inactivated after a few rounds of

Figure 4. Dosage compensation. The left part illustrates the problem with uneven X-

chromosome numbers between males (one copy) and females (two copies), and the uneven

ratio between the X-chromosome and the autosomes (two copies of each autosome) in

males. This is in mammals compensated by an inactivation of one of the female X-

chromosomes, by condensing it into a compact Barr body (small dot), but also by doubling

the expression of the X-chromosome in both sexes to equalize the X:A ratio. In Drosophila,

dosage compensation is mediated by restricting the doubling in expression of the X-

chromosome to males.

INTRODUCTION

21

cell division in the embryo [96], and condensed into a small, highly compact

element called the Barr body [97]. Females are thus mosaics for two cell

types that express one or the other X-chromosome. Only a few Xi-bound

genes escape the inactivation and are still expressed from the Barr body [96].

The X inactivation occurs early during embryogenesis and is initiated by

transcription of the 19 kb long non-coding RNA Xist, from an inactivation

center on the X-chromosome destined for inactivation. The Xist RNA

spreads in cis to cover the entire length of the X-chromosome [98,99] [100],

and initiates the silencing process, possibly by functioning as a platform for

recruiting repressive complexes (Polycomb complexes) [101]. Transgenes of

Xist placed on autosomes can induce spreading of Xist and different degrees

of gene silencing of autosomal genes surrounding the insertion site [102].

Evidence suggests that HP1 is involved in the process of inactivating the X-

chromosome in humans [103].

Up-regulation of mammalian X-chromosome

The inactivation of one of the female X-chromosomes solves the problem

with unbalanced sex chromosome ratio between males and female, however,

one additional problem remains with this model; the single active X-

chromosome in both males and females is out-numbered by the autosomal

chromosomes, which are all present in two copies. This leads to an

unbalanced X to autosomes ratio of 0.5-1, and the transcriptional output

from the single X-chromosome will thus be too small compared to the

transcriptional output from the autosomal chromosomes.

This problem appears to be solved by a doubling of the expression from the

X-chromosome in males and from the active X-chromosome in females

[102,104-106] (See figure 4). Up-regulation of mammalian X-chromosome

has been detected in several species including human, primates, rat and

mouse, and microarray studies have shown that the X:autosome

transcriptional output is close to 1 in most somatic tissues from both males

and females [104-107].

In fact, it is likely that up-regulation of the X-chromosome actually began to

evolve first, to balance the X to autosome ratio when the proto-Y in males

was gradually degenerated, and that the X-inactivation in females followed

as a response to the hyper-activation of the X-chromosome, which would be

unfavorable for females already possessing two X-chromosomes [108].

INTRODUCTION

22

Very recently, an additional long non-coding RNA, XACT, has been

discovered in a pluripotent human cell line. XACT coats the length of the

active X-chromosome in females as well as in males in a similar fashion to

the Xist covering of the inactive Xi-chromosome [109], indicating that this

non-coding RNA might be involved in the mechanism that up-regulates the

active X-chromosome in males as well as females. In absence of Xist, the

XACT RNA is expressed from, and covers the length of both X-

chromosomes.

However, it should be noted that mammalian X-chromosome up-regulation

is a controversial issue and some studies claims that up-regulation of the

mammalian X-chromosome does not exist at all [110].

In Drosophila

In Drosophila, dosage compensation is achieved by a two-fold up-regulation

of the male X-chromosome (see figure 4) [108], and this hyper-activation is

mediated at least to some extent by a male-specific ribonucleoprotein

complex, called the Male-Specific Lethal (or MSL) complex. The MSL

complex decorates the entire length of the male X-chromosome, by binding

to hundreds of distinct sites (shown by cytological studies on polytene male

X-chromosomes) (see figure 3B), and mediating H4K16 acetylation [111].

The proteins of this complex are, as implied by their name, essential for

viability in males but not in females and males lacking any one of the MSL

proteins will die during larval stage [95]. The complex consists of at least five

different proteins, MSL1, MSL2 and MSL3, MLE, and MOF and two non-

coding RNAs (ncRNAs), roX1 and roX2.

MSL1

MSL1 (Male-Specific Lethal 1), together with MSL2, appears to function as a

backbone for the assembly of the complex, and MSL1 can independently

interact with MSL2, MSL3 and MOF [112,113]. Evidence suggest that the C-

terminal domain (also called the PEHE domain) of MSL1 interacts with

MSL3 and MOF [114], and that a coiled-coil N-terminal domain of MSL1

interacts with a RING finger domain of MSL2 [115,116]. It has been

suggested that two MSL1 proteins initially form a dimer that is essential for

the MSL complex to assemble, recognize- and spread on the X-chromosome

[117].

INTRODUCTION

23

MSL2

MSL2 (Male-Specific Lethal 2) is the limiting component which stabilizes the

entire MSL complex; without MSL2 no complex is formed and in contrast to

the other components of the MSL complex, the MSL2 protein is only

expressed in males. The absence of MSL2 is also what prevents the MSL

complex from being formed in females, because the expression of MSL2 in

females is blocked at the level of translation by the female specific sex-lethal

(SXL) protein, thus, the remaining proteins expressed in females cannot be

assembled into a functional MSL complex. If ectopic expression of MSL2 is

induced in females, an MSL complex is formed and hyper-activates the

female X-chromosomes, leading to severely impaired viability [118-121].

MSL2 is a RING finger containing protein, meaning that it can potentially

bind to DNA [118,119,122]. An additional MSL2 regulating mechanism has

been proposed for male Drosophila, in which non-chromatin- associated

MSL complexes bind to msl2 mRNA in males. This blocks the translation

into MSL2 proteins and allows a fine tuning of the amount of functional MSL

complex, by feedback regulation of the rate-limiting component [123].

MSL3

MSL3 (Male-Specific Lethal 3) contains an N-terminal chromo domain, a

common domain within chromatin remodeling proteins [124], which is

needed for interacting with H3K36me3 on the nucleosomes of active genes.

MSL3 and MOF interact together and MSL3 must be acetylated by MOF at a

single lysine residue close to its chromo domain (at lysine 116) to be properly

included in the MSL complex [125].

MLE

MLE (Maleless) is an ATP-dependent RNA/DNA helicase, which means it

has the capacity to unwind DNA or double-stranded RNA by breaking the

hydrogen bonds between the two strands [126]. It is weakly associated with

the MSL complex [115,127], and the association of MLE with the X-

chromosome is sensitive to RNase treatment [128]. MLE interacts with the

roX RNAs, and this interaction appears to be necessary for both of them to

be incorporated into the MSL complex [129-131]. MLE has an ATPase

activity, which means it catalyzes the release of energy by de-

phosphorylating ATP into ADP, and this activity appears to be sufficient for

transcriptional activation, whereas the helicase activity is required for the

spreading of the MSL complex along the X-chromosome [132]. Additional

INTRODUCTION

24

function of MLE has also been proposed, in which it binds to newly

transcribed RNAs from the X-chromosome, and in this way it can direct the

MSL complex to active genes [128,133,134].

MOF

MOF (Males absent On the First) is a histone acetyltransferase (HAT) and it

specifically acetylates lysine 16 on histone H4 (H4K16), which is important

for the up-regulation of the X-chromosome [111,135,136]. The presence of

H4K16 acetylation does not appear to be directly involved in targeting or

spreading of the complex, since none of the components of the MSL complex

have any known domain for recognizing this modification. It thus seems

more likely that this modification facilitates spreading by opening the

chromatin structure, and thus, increasing the accessibility of the MSL

complex [137]. Unlike the other MSL proteins, MOF is encoded on the X-

chromosome.

MOF is the only MSL component that is also found associated with

autosomal chromosomes [137,138], and it appears to be part of a non-X-

specific complex called the NSL (Non-Specific Lethal) complex, which

targets promoters of constitutively expressed genes (housekeeping genes),

on both the X-chromosome and on autosomes in males and females. It

appears to be involved in the recruitment of RNA Pol II and the pre-

initiation complex to the promoters of the targeted genes [139,140].

However, in presence of the MSL complex, the catalytic activity of MOF is

mostly constrained to the X-chromosome [141].

roX1 and roX2

roX1 and roX2 (RNA on the X1 and 2) are two non-coding RNAs essential for

proper targeting of the MSL complex. They are both encoded on the X-

chromosome, at cytological section 3F3 and 10C7 respectively. These two

ncRNAs are very different in size and primary sequence, roX1 is

approximately 3.7 kb in length whereas roX2 is only 0.6 kb in length. They

only share one similar, 30 bp long sequence [142]. Despite these differences,

roX1 and roX2 have redundant functions, meaning that deleting one of these

two still creates a fully functional MSL complex, but deleting both is lethal to

males, with a few sterile escapers (in contrast to mutants of the other

components of the MSL complex which have no escapers). In these flies,

both the MSL complex and histone 4 acetylation are relocated to autosomal

INTRODUCTION

25

sites [143,144]. This creates interesting questions about the function of these

ncRNAs in the MSL complex.

The MSL complex binds to these two genes, and it has been suggested that

the MSL complex is incorporating the synthesizing roX RNAs at the site of

transcription, because autosomal transgenes of roX can induce spreading of

MSL binding at about a 1-2 Mb region surrounding sites of insertion

[145,146], and both roX RNAs are rapidly degraded if they do not associate

with the proteins of the complex [131]. Autosomal roX transgenes can also

rescue roX mutants which would otherwise die [144].

High affinity sites and spreading of DCC

It is known that the MSL complex binds to active genes on the male X-

chromosome, rather than over stretches of genomic regions (thus resulting

in the re-occurring banded pattern) (see figure 3B), and MSL1 and MSL2

forms the core of the complex that initially targets about 200 reproducible

sites distributed over the length of the chromosome. These sites have been

named High affinity sites (HAS) or chromatin entry sites (CESs)

[120,145,147-151] The MSL1- MSL2 core serves a platform on which the

additional proteins and ncRNAs assembles into a functional MSL complex,

which then can spread to cover the remaining non-CES targeting sites on the

X-chromosome. MSL3, MLE and MOF and at least one of roX1 or roX2 are

all required specifically for MSL spreading from the CES [144,149,151]. The

genes encoding roX1 and roX2 are furthermore two of the strongest CES for

the MSL complex and were the first CESs to be discovered [130]. CES

inserted into autosomal sites can induce short-range spreading to the 3’ end

of flanking genes surrounding the site of insertion [148].

Targeting mechanisms of the MSL complex

One very important question is how the dosage compensation mechanism is

able to distinguish the genes of the X-chromosome from the other

chromosome, and there are two main models for the targeting of MSL: the

transcription model and the sequence model.

The transcription model argues that MSL targeting depends on gene

activity, because the MSL complex is highly associated with active genes:

about 75% of all active genes on the male X-chromosome are bound by the

MSL complex; and about 80-90% of genes bound by the MSL complex are

associated with active chromatin marks, such as the transcription-associated

INTRODUCTION

26

H3K36me3 mark [146,152,153]. If the chromo domain of MSL3 is disrupted,

the MSL complex is not able to spread further than about 1 kb away from a

CES, which indicates that the recognition of H3K36me3 by MSL3 is involved

in spreading of the complex [154]. It is also known that the complex can

spread from an ectopic CES into active genes with no X-specific sequence

(i.e. translocation of CES to autosomes can induce spreading)

[145,146,154,155], and that autosomal genes inserted onto the X can be up-

regulated [156,157]. Also, non-dosage compensated X-bound genes can start

recruiting MSL upon activation [158]. Furthermore, it is interesting that even

though the dosage compensation seems to depend on H4K16 acetylation

mediated by MOF, many more genes (nearly all active genes) are acetylated

than are bound by the MSL complex [159] and more active genes also appear

to be dosage compensated than are bound by the MSL complex [105].

The sequence model on the other hand claims that the MSL complex

recognizes chromosome X specific sequences; sequence composition analysis

(i.e. searching for enrichment of specific types of sequences) of the X-

chromosome in Drosophila have revealed that the X is enriched for simple

repeat sequences, making it different from the autosomal chromosomes

[160,161]. Motif search in previously defined CES [162,163] has furthermore

identified a 21 bp long, GA-rich motif enriched in MSL binding, this motif is

named MSL recognition site (MRE) [148]. Sequence motif have also been

suggested by [153,164], although they do not perfectly predict the MSL

binding.

However, it is likely that the targeting mechanism is actually a combination

of these two models, so that the MSL complex first recognizes specific

sequences in the CES, and then spreads to nearby genes based on chromatin

marks for active transcription [146]. It has also been suggested that the

secondary spreading mechanism as well could depend on sequence-

specificity [165].

Mechanism behind the X-chromosome up-regulation

The MSL complex binds mainly to the gene body of active genes, with a

preference for the middle part and the 3’ end of the genes [152,153]. In

contrast to many transcriptional regulators, it does not bind to promoters.

The 3’ end binding bias is observed for many factors that regulate

transcription elongation, and this suggests that the MSL complex is

important for functions such as transcriptional elongation or recycling

of the polymerase back to the promoter [152,153] [166]. However, recently

INTRODUCTION

27

this 3’ bias was suggested to depend on the fact that the 3’ end usually

contains longer exons, and that the MSL complex rather binds with an exon

bias than a 3’ bias [147]. The H4K16 acetylation mediated by the MSL

complex on the X-chromosome is also said to be responsible for hyper-

activation by loosening up the chromatin structure and thus allowing easier

transcription initiation [167], or enhanced transcription elongation [168].

However, it is relevant to keep in mind that although the X-chromosome in

males is hyper-activated about two times, i.e. has an expression level in

range with that of the autosomes, the up-regulation mediated by the MSL

complex actually only counts for about a 1.4-fold hyper-activation [169-173].

This indicates that another compensating mechanism exists as a

complement to the MSL complex.

PAINTING OF FOURTH (POF)

For a long time, the concept of chromosome-specific gene regulation was

ascribed only to sex chromosomes and dosage compensation mechanisms;

however, with the discovery of a particular protein that seemed to only bind

to the small 4th chromosome in Drosophila, this view was challenged. This

protein was named Painting of Fourth, or POF [174]. POF binds to the

distal section of the fourth chromosome and is so far the only described

chromosome specific protein that binds to an autosome (see figure 3B). The

Pof gene is located in region 60E on chromosome 2R and it encodes a 495

amino acid long protein containing a predicted RNA-binding domain

(RRM1) in the central part of the protein [174].

Specifically, POF targets the gene body of active genes on the 4th

chromosome and has a strong preference for exons, with a tendency for a

stronger binding at the 3’ end [46,48,175], and it is responsible for

stimulating expression to similar levels as the MSL complex. A Pof mutant is

viable but has a significant general reduction of chromosome 4 gene

expression [48].

The binding of POF is quite specific for the 4th chromosome because partial

translocation of chromosome 4 onto the tip of other chromosomes cannot

recruit POF, nor can POF binding spread into regions of other autosomes

that are translocated onto the tip of the 4th chromosome [174].

INTRODUCTION

28

Chromosome 4

The 4th chromosome of Drosophila is unique and differs from the other

chromosomes in several ways; the most obvious difference is the size; the 4th

chromosome is very small, its total length is estimated to be 4.5-5.2 Mb or about 3.5% of the genome (see figure 3A) [176]. Secondly, even though it

contains about 92 genes, it is considered to be a mixture of both euchromatic

and highly heterochromatic regions. The entire 4th chromosome displays

heterochromatic features such as late replication [177] and no meiotic

recombination, and the proximal part of this chromosome, which also

constitutes the majority of the length (about 3-4 Mb), is a highly condensed,

completely heterochromatic structure. The proximal region is under-

replicated in polytene tissue (i.e. does not go through endoreplication and is

therefore not visible in polytene chromosome), mostly consists of simple

satellite repeats and does not contain any known genes [178].

The remaining 1.23 Mb of the 4th chromosome constitutes the polytenized

distal part, which roughly corresponds to the banded region visible in

polytene chromosomes (cytological sections 101E-102F). The 92 genes

associated with the 4th chromosome are all found within this banded section,

giving it a gene density close to that of the major chromosome arms.

However, this region also has several properties typical of heterochromatin:

the genes are interspersed with a mix of unique sequences, repetitive- and

transposable elements [179-182], and it is associated with the

heterochromatin inducing proteins HP1a, SETDB1 and the histone

modification H3K9me. It also has the capacity to induce variegated

repression on inserted reporter genes in a way that is typical of

heterochromatin position-effect variegation (PEV) [183].

So this gene-containing section of the chromosome is actually a mixture of

euchromatin and heterochromatin [180], and this means that the active

genes on the 4th chromosome have to be able to uphold expression within a,

in many aspects, repressive environment [184,185]. It is even so that the

ability of POF to bind chromosome 4 depends on this heterochromatic

background; increasing the compaction of heterochromatin, by decreasing

the temperature or removing the Y-chromosome, can results in binding of

POF to a translocated 4th chromosome [48]. This makes the 4th chromosome

a very interesting model for studying heterochromatin and its effects on gene

expression.

INTRODUCTION

29

Haplo-4th lethality and POF

One very interesting feature of the 4th chromosome is that it is the only

autosome in Drosophila that can exist in only one copy, without causing the

flies to die, i.e. it is haplo-viable [178]. But if you combine a haplo-4th

condition with a Pof mutation (which on its own does not cause lethality),

the flies apparently can no longer sustain sufficient chromosome 4

expression and die. Therefore, it seems as if POF is essential for maintaining

chromosome 4 gene expression in flies which are deficient in chromosome 4

dose [48].

Balanced regulation of chromosome 4 genes by POF and HP1a

Interestingly, the binding of the stimulating protein POF and the repressing

protein HP1a to the distal, euchromatic part of chromosome 4 genes overlap

almost perfectly and they both bind with a preference for exons of the active

genes [45,46,48]. Furthermore, the binding of the two proteins appears to be

interdependent on each other; if HP1a is absent, POF is released from the 4th

chromosome and if POF on the other hand is absent, the HP1a binding to the

4th chromosome is reduced; although, this reduction is less pronounced [48].

This was, however, contradicted by a recent study showing that POF binding

is essentially unaffected in an HP1a mutant [54], but either way, they still

bind to essentially the same genes, and the opposing effects on gene

transcription suggest a system in which the expression of chromosome 4

genes is fine-tuned by a balancing mechanism between HP1a and POF, in

which POF provides a stimulatory function and HP1a provides a repressing

function [48].

Importantly, there is an exception to the overlap between POF and HP1a

binding to the distal section of 4th chromosome, and that is the

uncharacteristic H3K9me-independent HP1a binding to the promoters,

which are not targeted by POF [45,46]. In line with the fact that POF only

correlates with H3K9me-dependent HP1a binding; POF is also suggested to

interact with SETDB1, and the H3K9 methylation pattern on chromosome 4

is impaired in a Pof mutant, suggesting that the proper function of SETDB1

depends on the presence of both HP1a and POF at the gene bodies [73].

Evolutionary links between POF and the MSL complex

The apparent requirement of POF to maintain sufficient chromosome 4 gene

transcription in a haplo-4th condition gives POF a whole-chromosome

INTRODUCTION

30

regulatory function, reminiscent of the dosage compensation system of the

male X-chromosome. The stimulating potential of POF and the MSL

complex are also comparable to each other. But there are also other

evolutionary aspects connecting POF and chromosome 4 to the MSL

complex and the X-chromosome: the binding of POF to chromosome 4 (also

known as the F element) is evolutionary conserved in several different

species of genus Drosophila (for example in D. virilis which diverged from

D. melanogaster about 39 million years ago [186]). And in some species, like

D. ananassae and D. malerkotliana, POF in fact decorates the entire male

X-chromosome and co-localizes perfectly with the MSL complex.

Furthermore, in D. busckii, where POF not only binds to the X-chromosome,

but where the entire F element is actually fused to the X-chromosome, there

is no apparent MSL complex (i.e. no complex that is detectable with a D.

melanogaster MSL-antibody), instead, POF is decorating the X-

chromosome in males, but not females, and this binding co-localizes with

H4K16 acetylation (which is characteristic for D. melanogaster dosage

compensation of the male X). This suggests that POF and the MSL complex

are evolutionary connected [187]. It is noteworthy that of the tested relatives

of D. melanogaster within genus Drosophila, D. busckii is also the most

distant relative that contained POF, and this suggests that POF might have

originated as an early form of dosage compensating mechanism for up-

regulating the male X-chromosome [46,48,172,174,187,188].

Moreover, in similarity with the X-chromosome, chromosome 4 seems to

have “female tendencies”, meaning that in intersexes with a 2X:3 autosome

ratio, an increase in chromosome 4 dose will lead to female development,

whereas a decrease in dose will lead to male development [189,190].

Furthermore, three copies of the 4th chromosome increase the frequency of

chromosome X nondisjunction, which indicates that chromosome 4

sometimes pairs with the X-chromosome during meiosis [191].

Do other compensating systems exist?

I have discussed the highly evolved mechanisms for compensation of sex

chromosomes that aim to equalize expressional output between the X-

chromosomes of XX:AA females and X:AA males, as well as the equalization

between the autosomes and the single X-chromosome ratio in mammals.

POF, and the up-regulation of chromosome, 4 presents the so far first

described autosome specific regulating protein, and this facts gives rise to

the perhaps not very far-fetched question: are there other similar

INTRODUCTION

31

mechanisms for compensating un-balanced autosomal regions in general?

Unbalance in genome copy number is generally referred to as aneuploidy,

and this is not at all an uncommon phenomenon in nature, so aneuploid

genomes can provide good models for investigating autosomal compensatory

mechanisms. I will first explain the concept of aneuploidy, and its general

effects and consequences:

ANEUPLOIDY

Aneuploidy and segmental aneuploidy refers to a state in which one or

more chromosomes or chromosome segments are present in abnormal copy

numbers. The normal chromosome copy number for many species, including

mammals and Drosophila, is two, meaning that all chromosomes (and

hence, all genes) are present in two essentially identical copies, one set is

inherited from the mother and one set is inherited from the father. The

exception is the sex chromosomes, which are unequally distributed between

the sexes. Aneuploidy is defined as loss or gain of one or more (but not all)

chromosomes, giving a chromosome number that is not an exact multiple of

the haploid number. This is slightly different from polyploidy, in which the

cell or organism carries three or more complete sets of chromosomes. This

phenomenon is frequently found in nature as the normal chromosome set up

of many plants and some animals, including a few specific human cell types.

Aneuploidy usually occurs during meiosis: Normally, when chromosomes

are being duplicated, the two copies of the chromosomes are equally

distributed to the two daughter cells, so that the total number of

chromosomes is maintained in the cell line. But if you have errors in the

chromosome segregation, chromosome copies can be lost or an extra copy

can be obtained (aneuploidy states which more specifically are termed

monosomy and trisomy, respectively). If the segregation problem is

combined with double strand breaks of the chromosome, it might lead to

segmental aneuploidy.

Cancer, developmental diseases and aneuploidy

Aneuploidy in humans in most cases lead to fetal abortion at early

developmental stages and is actually the leading cause of miscarriages in

humans [192,193]. It is predicted that over 10% of human oocytes are

aneuploid, although these oocytes rarely give rise to viable offspring [194].

However, in cases where the offspring is viable, the copy number defects

always give rise to developmental abnormalities and a reduction in fitness

INTRODUCTION

32

occurs in all species studied [193,195-200]. Most mental retardations in

humans are caused by aneuploidy [196,201], for example trisomy 21 (Down

syndrome) which is an extra copy of the relatively small chromosome 21 in

humans. As can be expected, the severity of the associated defects of

surviving individuals with aneuploidy can to some degree be correlated with

the length of the aneuploidy region, and in general, segmental trisomies are

better tolerated than segmental monosomies [195,197,200].

In addition to developmental abnormalities, aneuploidy is one of the

hallmarks of cancer, and tumors often contain extensive aneuploidy, with

gain or loss of multiple chromosome copies or segments of chromosomes

[201]. This is very contradictory with the nature of aneuploidy, since this in

normal cases is deleterious to a cell and to organisms as a whole. So the fact

that tumor cells survive these deleterious chromosomal rearrangements has

remained a mystery, but it also remains very complex and controversial

whether aneuploidy is a cause or a consequence of cancer [195,200]. Either

way, the aneuploid state, which would normally be deleterious, is clearly

accepted in tumor cells since these have gained proliferative advantages

compared to normal cells within their surroundings. This provides

interesting future questions to understand how or by which mechanisms

tumor cells are able to evade the deleterious effects of abnormal copy

numbers, and consequently the unbalanced levels of gene products.

Aneuploidy and evolution

One important aspect to consider in aneuploidy is that although an

aneuploidy state is usually deleterious to a cell, and in theory should be

selected against, variations in the number of chromosome or chromosomal

segment copies is important for genomic variance and hence in the long run,

evolution. With this in mind it is understandable that cells and organisms

need to be able to survive certain amount of copy number variation to

potentially obtain new traits, and will thus have evolved mechanisms for

this. These mechanisms need to be balanced between being efficient enough

so that the cells survive minor copy number variation, but without

compensating away the potential beneficial effects [172,199]. It turns out

that differences in copy number are relatively common in all individuals of

all species studied. About 500 individual regions or around in total 0.2% (6

Mb) of the genome may for example vary in copy number between two

normal human individuals [202,203], and recent studies have shown that up

to 13% may vary in copy number between two individuals [199], indicating

the importance of aneuploidy for creating genomic variation.

INTRODUCTION

33

Aneuploidy in Drosophila

The Drosophila genome has been divided into 102 numbered divisions based

on cytological analyses of the polytene chromosome [204], and each region

is about 800-1500 kb in length [205]. Induced monosomic regions (i.e one

copy number) longer than 800-1500kb, or more than 1% of the genome

reduces fertility and viability [197]. Although, there are a couple of

exceptions to this rule, where flies carrying a ~1.7 Mb or a ~2.8 Mb

monosomic region respectively, are still viable [206]. In general, Drosophila

are more sensitive to segmental monosomy than trisomy (in the whole

genome so far only one triplo-lethal allele and two alleles that give mutant

phenotype in triplo condition are described [197]). Still, only about 50 out of

all the approximately 15 000 genes in Drosophila are haplo-insufficient,

meaning that they are lethal when one of the copies (alleles) are mutated.

One large class of these haplo-lethal loci is the “Minute” genes. Monosomic

state of these loci results in flies with short, thin bristles, reduced

developmental rate, low viability and fertility and other abnormalities

[207,208]. This means that for the large majority of genes, one wildtype gene

copy is sufficient for proper function of the organism.

Are there mechanisms for aneuploidy compensation?

The various aspects of aneuploidy listed above hints that there should be

some mechanism involved, which can diminish the deleterious effects of

autosomal copy number variation and allow cells to survive minor

rearrangements. Also, that the MSL complex only accounts for about 1.4

times of the up-regulation of the two-fold hyper-acetylated male X-

chromosome [171] indicates the existence of another, more elusive

compensating mechanism.

Buffering

Indeed, several studies have shown that autosomal aneuploidy regions are

actually affected by an elusive mechanism which compensates the expression

level of these regions, both up or down, and this autosomal compensation

has been named buffering (see figure 5).

When considering aneuploidy, i.e. variations in the dose of a genomic region

and the effects on expression level, you would intuitively expect that if a

genomic region was present in either one copy (monosomy) or three copies

(trisomy), the transcriptional output from this region would drop to 0.5 or

INTRODUCTION

34

increase to 1.5 respectively, compared to a normal two copy region where the

transcriptional output would be 1. However, it has been shown in a number

of early studies in segmental trisomies, that duplicated genes had a lower

expression output than expected, indicating that something is reducing, or

buffering, the expression of the three autosomal gene copies, and this was

measured both on protein level [209-211] and RNA level [209,212]. Studies

using more genome-wide approaches, such as microarray, supported these

findings and showed that buffering appears to exist also on a wider scale

[105,213-216]. For example one of these studies performed in humans found

that triploid regions were expressed at a mere 1.1-fold, compared to the

expected 1.5-fold, indicated that the triploid regions are repressed to obtain

an transcriptional output level more similar to that of a wildtype [215]. Two

studies also showed that a three-fold gene dose difference (i.e. when

comparing the expression from three copies of an autosomal region to that of

one copy number of the same region) only results in a 1.4-fold change in

mRNA levels [105,216]. Thus, it is evident that autosomal compensation, i.e.

buffering exists, albeit elusive.

Buffering could explain why the male X-chromosome has a two-fold up-

regulation even though the MSL complex only accounts for a 1.35-fold up-

regulation [171]. It was speculated by [105,108,171,172] that if we assume

that buffering stimulates monosomic regions to an average of an 1.5-fold up-

regulation and that it can act on all chromosomes with a deviating copy

number, including the male X-chromosome, then buffering together with the

effects of the MSL complex can count for the two-fold hyper-activation of the

male X-chromosome, 1.50-fold x 1.35-fold=2.03-fold!

Even if this particular effect of buffering of the X-chromosome can be a bit

hard to measure exclusively, the conclusion is that autosomal buffering does

exist to allow cells and organisms to survive milder forms of chromosome

imbalance, and it appears to act on the transcript level. However, the

buffering effects do not fully restore transcript levels back to a wildtype

(diploid) level, and it is therefore possible that buffering occurs on more than

one level. A study in a number of yeast strains, carrying extra chromosome

copy numbers, showed that mutations leading to increased proteasomal

degradation had evolved to increase the fitness of several trisomic yeast

strains [217]. This suggests that buffering is something that also acts on a

protein level, and thus it is obvious that many aspects about buffering

remains to be explained.

INTRODUCTION

35

One important question is how this buffering mechanism works; and also

what causes the haplo-insufficiency once the aneuploidy region becomes too

extended: either it could be a general mechanism that recognizes and targets

monosomic/duplicated regions and stimulates or inhibits gene expression

more or less evenly over these regions, and that the cause of haplo-

insufficiency is depleted levels of the buffering components. This model is

supported by the fact that when the size of an aneuploidy region reaches a

Figure 5. Schematic illustration of buffering. A normal disomic region in two copies

(one copy inherited from the mother and one copy inherited from the father) compared to a

monosomic region (i.e. a region that has lost one copy) with and without buffering, and a

trisomic region (i.e. a region with one extra copy) with buffering. In the absence of any

buffering, the expected transcriptional output of the monosomic region is 0.5 of that of the

disomic region. But if the monosomic region is buffered, an unknown mechanism induces a

higher expression of the remaining single copy genes so that the total amount of

transcriptional output is closer to that of the normal disomic level. In a trisomic region,

buffering works in the opposite way, repressing gene expression to reduce the amount of

transcriptional output.

INTRODUCTION

36

critical point, it is primarily the extent of the aneuploidy that causes the

lethality and not the uncovering of specific genes or regions [197]. The other

option is that buffering is the sum of regulation of individual genes, which

would mean that most genes within a deficient region are unaffected

whereas a few genes become fully compensated by feedback regulation. The

haplo-insufficiency would then primarily be caused by collapsing gene

networks. So as long as a deficiency is small enough, the probability is that

the affected genes are functionally unrelated to each other, and the

remaining components of the protein network can balance for the lack of one

network component. As the deficient region expands, you will eventually

reach a point in which more than one gene involved in the same protein

network is affected, and consequently, the other components of the network

can no longer cover up the losses, resulting in lethality [218].

It addition to the putative buffering mechanism or mechanisms, other less

general mechanisms could also be involved in gene dose regulation, such as

feedback regulation and feedforward regulation.

Feedback regulation

Feedback regulation is a mechanism for controlling gene expression by the

level of gene product present. If levels of a protein or RNA are too low, gene

product is continuously produced until sufficient levels are reached, and the

newly synthesized proteins will inhibit the formation of more proteins. This

makes feedback regulation a well-established error control mechanism

where the protein itself regulates the amount of new protein produced.

Feedforward regulation

Feedforward mechanism on the other hand functions by anticipating

possible effects of system errors rather than acting once the levels are out of

balance. For example the cells can detect copy number variation and adjust

transcription levels prior to major errors occur.

One specific example is the sex determination in male Drosophila embryos:

the cells detect the X copy number in relation to autosomal copy number and

assigns the MSL complex to the X before zygotic transcription is activated in

a feed-forward manner [219]. Although the MSL actually also displays

feedback properties by binding to transcribed genes [152,163,220,221].

INTRODUCTION

37

Inverse dosage effect

An alternative view on aneuploidy compensation is that since most gene

regulatory enzymes are repressing, loss of a chromosomal segment would

lead to a greater loss of negative regulators than of positive regulators, and

consequently the remaining genome will increase in transcription, rather

than the monosomic region being decreased in expression. According to this

model, the monosomic state of the X-chromosome in males causes a general

up-regulation of the entire genome and the MSL complex acts by attracting

and isolating the activating factor MOF to the X-chromosome, and hence

passively causing an up-regulating of the male X-chromosome [222]. This

model is however contradicted by the finding that an msl2 mutant causes a

reduction of chromosome X expression, but with very little effect on the

autosomes [169,223]. Yet, the true reasons behind these differences in X-

chromosome and autosomal expression levels are very hard to determine

and this issue presents one of the most problematic aspects of measuring

genome-wide effects of gene expression:

Challenges with genome-wide expression analysis

Reference points

In studies involving gene expression effects of chromatin-binding proteins,

chromosome copy number variation, or other conditions in general that

affect large parts of the genome, one key concern for accurate estimations is

to set valid reference points, and this is a general problem in these kinds of

studies. If you measure effects of a few genes, which most likely will have

very minor effects on the remaining genome, the whole genome or a few

reference genes can be used as the standard to which the effects of your

genes of interest can be compared to. But if you start applying severe

conditions to the cells, or study aneuploid regions that are big enough to

affect the rest of the genome, it becomes very difficult to set a reference

point, to what should you compare the effects that you are interested in?

Skewness

Another related issue is skewness since it is evident that different data set

will always be affected by both biological and technical variation. Two

replicates that are generated by genome-wide expression arrays (such as

next generation sequencing or microarrays) can never be directly compared

as raw data, instead they need to be normalized against each-other so that

INTRODUCTION

38

they obtain in principle the same array average value, and the same variance.

If gene expression of a large part of the genome is affected, for example by a

down-regulation of the X-chromosome, the normalization process will by

default try to equalize the overall average expression level of the whole data

set, and this will mean that the down-regulation of one large region will be

compensated by increasing the expression value of the remaining genes, thus

creating a bias [224]. This will have the result that it is very hard to

determine whether a change between the X and the autosomes is caused by

just a down-regulation of X or also by an up-regulation of the autosomes.

Thus, the challenge with genome-wide studies is keeping a balance between

measuring the effects of large genome rearrangements and still avoiding

biased data sets.

This is one of the reasons why the small 4th chromosome is such a good

model for studying chromosome-wide gene regulation; it has its own unique

regulatory mechanism (POF) and contains enough genes (92) to measure

significant effects, but it is still small enough not to affect the remaining

genome, and thus it is easy to set reference points. The same is true for

segmental aneuploid regions (such as those used in paper I and II in this

thesis). They are in approximately the same size range as chromosome 4, and

will thus be very useful in examining effects on copy number variation and

the potential autosomal buffering effect.

Limitations in the arrays

An additional problem with several genome-wide methods, especially

microarrays, is that there are limitations in the detection range. The

expression levels of genes that are expressed at very low levels (i.e. un-

expressed) will inevitably drown in background noise of the microarrays,

meaning that any differences in transcriptional output between genes in a

monosmic or trisomic state vs. a diploid (wildtype) state will not be detected,

and these genes will therefore appear as if they are fully compensated. This is

a general problem with many of the previous estimations of buffering level,

which have usually included all genes, both expressed and unexpressed.

Therefore, these studies may tend to overestimate the buffering effect, and

this is a problem we have addressed in paper I and II.

Aims

39

AIMS

The general aims of this thesis have been to:

Investigate the buffering effect of segmental monosomic regions and

find out if there are general responses in the genome that are

triggered by aneuploidy (Paper I and II).

Determine whether buffering is a general mechanism or a gene

specific mechanism, and potentially find features of genes and

regions which affect the level of buffering (Paper I and II).

Elucidate the conflicting reports about HP1as function on gene

expression; inhibiting or stimulating? (Paper III).

Determine which roles the two different HKMTs, SETDB1 and

Su(var)3-9 are playing on gene expression in different parts of the

genome (Paper III).

Use POF binding sites located on the female X-chromosome to

determine the relationship between POF and dosage compensation,

and furthermore, determine how POF binding is influenced by

chromatin environment (Paper IV).

RESULTS AND DISCUSSION

40

RESULTS AND DISCUSSION

PAPER I AND II

General buffering levels

In segmental monosomic regions, we found that the transcription output

was indeed enhanced in relation to the gene dose, i.e. buffered. The average

expression varied depending on region between 54%-58% of wildtype

expression in paper II, and up to 64% of wildtype expression level in paper I.

If no buffering effect existed, we would expect these regions to be expressed

at 50% of wildtype expression, corresponding to the 0.5 gene dose. Our

measured buffering effect is in general lower than previous studies have

shown [105,213,215], however, we postulate that this is caused by the more

stringent cutoff we applied, excluding all genes expressed below or above the

reliable detection range of the array. The excluded genes are in the risk of

being scored as fully compensated if the array, due to background noise or

over-saturation, is unable to detect differences in gene expression between

the aneuploid and the control samples. We conclude that buffering of

monosomic regions is a weak but significant effect, which acts to stimulate

expression of the remaining single copy genes to compensate for the reduced

gene dose. Furthermore, we find that the average buffering of entire

monosomic regions is weakly related to the number of genes that are in

single copy, the more genes affected, the lower the buffering is. Yet, buffering

is not clearly decreased if another monosomic region is added in the same

genome, nor could any other clear effects be seen when combining

monosomic regions.

Buffering of specific gene groups

On average, when studying compensation of whole regions, our estimated

buffering effect is small but significant, however, when we start looking at

specific groups of genes, the buffering effect becomes evident. We found that

buffering depends on expression pattern, wildtype expression level, and gene

length:

UEGs and NUEGs

We found that in monosomic regions, Non-Ubiquitously Expressed

Genes (NUEGs) (i.e. genes which are only expressed in some, but not all

Results and discussion

41

tissue types of the body) are more strongly buffered than Ubiquitously

Expressed Genes (UEGs) (i.e. genes which are expressed in all tissue

types). UEGs are likely important for maintaining general housekeeping

functions and need to be expressed at relatively high levels in all tissue types,

whereas NUEGs on the other hand are required for more tissue specific

functions in general, and are consequently only active in some tissue. We

therefore speculate that more regulatory mechanisms have evolved to fine

tune expression of NUEGs than UEGs. UEGs are already expressed close to

the maximum levels, and therefore, it is hard to further stimulate these

genes, whereas NUEGs still have the potential to have their transcription

enhanced in most tissues. One could even speculate whether UEGs are

buffered at all; it is possible that the weak effects measured for UEGs are

mainly due to misclassified NUEGs, which is likely to occur since

classifications such as the one used will never be perfect.

Gene length and wildtype expression level affects buffering

When looking at the buffering of individual genes, the strongest determinant

for buffering we found was gene length: long genes (>3 kb) display the

strongest buffering, and for these genes, the buffering effect is also

independent on wildtype expression level or expression pattern (i.e. NUEG

or UEG). Short genes (<3 kb) are less buffered on average, but interestingly,

these genes in contrast actually depend on both expression pattern and

expression level, with a low expression level resulting in nearly the same

buffering as long genes. The short genes with highest expression level are not

buffered at all, and virtually the same is true for short UEGs. Thus, gene

length appears to be the primary determinant for buffering (see figure 6).

Buffering mechanisms

One very interesting and important question concerning buffering is: what is

the underlying mechanism behind buffering? Is it the sum of feedback

regulation of a few individual genes, or in the form of protein network

effects? Or is it a more general mechanism that recognizes and acts on

regions present in only one copy?

If the buffering was caused by feedback regulation, we would most likely see

a strong buffering effect only in a few monosomic genes, whereas the

majority of genes are expressed at around the expected 50% of wildtype. In

contrast, we observe a normal distribution of the buffering, on average most

monosomic genes are up-regulated and the variation is likely caused by

RESULTS AND DISCUSSION

42

normal variance and array noise. This indicates that buffering is mediated by

a more general mechanism, although, there is also a possibility that the

normal distribution of the buffering is caused by feedback regulation of

specific genes and that the buffering effect spreads, since spreading is a well-

known mechanism in, e.g., dosage compensation [130,145,152,225].

However, we found that buffering does not spread to neither neighboring

genes within the monosomic regions, nor to neighboring diploid regions,

which argues against a spreading mechanism.

The finding that gene length is the primary determinant for buffering makes

it tempting to speculate that transcriptional elongation is involved in this

mechanism. Transcriptional elongation has been proposed as mechanism for

the MSL complex [168,226], and considering the hypothesis that total

compensation of the male X-chromosome is a combined effect of the MSL

complex and a more elusive general mechanism, such as the autosomal

buffering, makes this even more tempting. However, it remains to be

clarified whether transcription elongation is really a part of the buffering

mechanism. Buffering could also be mediated, or partially mediated, by

looping out effects, since the monosomic regions are unpaired, they might

loop-out from the normal nuclear position into more permissive

environments, or by increasing access of transcriptional machinery.

POF compensates chromosome 4

Buffering could potentially also be mediated by a specific factor, similar to

the chromosome 4 specific protein POF. Chromosome 4 is the only autosome

which is haplo-viable, meaning that flies can survive with only one copy of

chromosome 4 [178]. In paper I, we found that a haplo-4th chromosome is

buffered to 70% of wildtype expression, which is a more pronounced effect

than on the other autosomes. Since we know that haplo-4th flies dies in the

absence of POF [48], we speculate that POF is involved in the buffering of

chromosome 4 genes. In paper I, we further support this hypothesis by

showing that there is a negative correlation between gene expression in a

haplo-4th condition and a Pof mutant. Genes that are most tolerant to loss of

one copy of the 4th chromosome are also most down-regulated in a Pof

mutant, and reciprocally, the genes that are not compensated by POF are

also the genes which are most sensitive to loss of one copy of chromosome 4.

In general, it appeared as if the genes compensated by POF were

overrepresented by NUEGs. We conclude that POF is the mechanism that

compensates chromosome 4 genes and that it preferentially stimulates

expression of NUEGs.

Results and discussion

43

Buffering induces proteolysis

We were interested to find out if any general effects were induced elsewhere

in the genome (i.e. in diploid genes not uncovered by any copy number loss),

as a general response to deletions, and we discovered that genes involved in

peptidase and proteolytic activity where highly over-represented among the

100 most up-regulated genes in the genome. Furthermore, we found a

negative correlation between buffering level and expression of proteolytic

genes: the poorer buffering a monosomic region displayed, the more

expressed the proteolytic genes were. The correlation was stronger for short

genes, in line with the previous finding that buffering is lower for short

genes. Proteolytic proteins are involved in the degradation and digestion of a

vast number of proteins, and are responsible for degrading misfolded

proteins, thus giving them an important regulatory function. Other studies

performed on duplications, i.e. genome containing extra copy number, have

also shown induced proteotoxic stress [217,227,228]. It is possible that

proteolysis is a general response to unbalanced genomes and is induced to

compensate unbalanced genomes on a protein level, when the buffering on

transcription level has not been sufficient.

Future perspectives

The studies of buffering included in this thesis are all conducted on fly

strains in which the deletion of a genomic region was induced long ago,

meaning that many generations have passed since the induction, and this

inevitably includes the possibility that the effects we observed have occurred

over long time as an adaptation to these impairments. Therefore, one

relevant question to be addressed is: would the same effects be obtained if

these deletion where re-induced, and the effects monitored directly upon

induction? What would happen if the buffering effects where followed over

several generations?

RESULTS AND DISCUSSION

44

Conclusions

Buffering of monosomic regions is a weak but significant effect,

ranging between about 54-64% of wildtype expression level.

Buffering of entire regions partially depends on the number of genes

included in the monosomic region, where regions with many genes

are less buffered.

Buffering of one monosomic region does not affect the buffering of

another monosomic region, if they are combined in one genome.

Gene length is the primary determinant for buffering. Short genes

depend on expression patterns and expression level for proper

buffering, and short UEGs and short genes with a high expression

level are not buffered at all (see figure 6).

Buffering effect is normally distributed, indicating a general

mechanism rather than feedback regulation of individual genes, and

buffering does not display any spreading to nearby genes.

POF is responsible for the compensation of haplo-sensitive genes on

chromosome 4, and is also crucial for the survival of haplo-4th flies.

Aneuploidy in general induces expression of proteolytic genes, and

this effect is more pronounced in genomes with poor buffering.

Figure 6. Schematic illustration of genes in a monosomic region. Buffering is

strongest for long genes, irrespective of gene type, whereas buffering of short genes

depends on gene type. Short UEGs are weakly or not at all buffered.

Results and discussion

45

PAPER III

HP1a has opposing functions on chromosome 4 and in

pericentromeric regions

It has been debated whether HP1a has a repressive or a stimulating function

on gene expression, and to better understand the function of HP1a, as well as

that of the co-dependent proteins POF, SETDB1 and Su(var)3-9, we analyzed

mutants of each of these genes using microarray expression analysis. The

most striking finding was that HP1a inhibits gene expression on the 4th

chromosome and stimulates gene expression in the pericentromeric regions

(the heterochromatic regions adjacent to the highly heterochromatinized

centromeres). To add even extra complexity to this matter, we discovered

that these effects where different depending on gene type: An HP1a mutant

was primarily affecting NUEGs in chromosome 4, whereas in the

pericentromeric regions, only UEGs were significantly affected. These two

regions are very different in features and genomic environment which

indicates that HP1a have evolved to adapt to different chromatin

environments. This could also explain the contradictory reports about the

function of HP1a.

HP1a has different functions at the promoter and at the gene body

We know that HP1a binds to active genes on chromosome 4, and that it has a

stronger binding peak to the promoters than to the gene body of many of the

genes. When re-analyzing this HP1a binding data [45], we found that the

binding is elevated to both NUEGs and UEGs of chromosome 4 and the

pericentromeric regions, compared to a control region. Interestingly, we

found that there is a difference between the gene types; HP1a binds much

stronger to the promoters of UEGs than the promoters of NUEGs, and

furthermore, there was also a region specific difference; HP1a targets the

gene bodies of chromosome 4 UEGs stronger than the gene bodies of

pericentromeric UEGs. This is very intriguing considering that HP1a binding

to gene bodies of chromosome 4 is dependent on H3K9me2 and me3, (and

thus indirectly dependent on SETDB1) whereas the binding to promoters is

independent on H3K9me2 and me3. It has been suggested that HP1a

binding to the promoters results in an open chromatin conformation which

stimulates gene expression, whereas HP1a binding to gene bodies has a

repressive function. The H3K9me-independent binding to the promoter is

likely also more stable and less transient than the binding to H3K9

methylation at the gene body [45,229].

RESULTS AND DISCUSSION

46

We therefore propose a model in which HP1a has adopted its function to

different genomic environments and that this is mediated through different

binding mechanism (see figure 7): in UEGs, HP1a preferentially binds to the

promoters and stimulates expression, maybe by loosening up the chromatin

structure. In the pericentromeric regions, the strongest HP1a binding is

found at the promoters of UEGs and this would also explain both the average

stimulating effects of HP1a in these regions, and the fact that only UEGs are

significantly affected in an HP1a mutant. On chromosome 4, HP1a binding is

also enriched at the gene bodies, and this repressive binding balances the

stimulating promoter binding of UEGs, causing these genes to be less

affected in an HP1a mutant. The NUEGs on the other hand, which lack the

promoter peak, are dominated by the repressive effect of HP1a at the gene

bodies, and are thus up-regulated in an HP1a mutant. Interestingly, we show

in paper I that POF preferentially stimulate gene expression of NUEGs,

indicating that POF might have the role of balancing the repressive influence

of HP1a in those genes that lacks HP1a at the promoters. The hypothesis that

the repressive effect of HP1a correlates with H3K9me is also supported by

our finding that HP1a and Su(var)3-9 to a high extent appear to repress

expression of transposons in an overlapping manner.

SETDB1 and Su(var)3-9 are complementary to each other

SETDB1 mediates the H3K9me2 and me3 on chromosome 4, which repress

chromosome 4 gene expression, and in paper III we found that SETDB1 co-

localizes with the POF and HP1a binding regions on chromosome 4, and that

Setdb1 and HP1a mutants cause approximately the same up-regulation of

chromosome 4 gene expression, with a preference for NUEGs. This supports

the fact that SETDB1 is essential for the HP1a binding to gene bodies on

chromosome 4. The Su(var)3-9 mutant background resulted in down-

regulation of genes in the pericentromeric regions, in a similar manner as

the HP1a mutant. This is surprising since the function of Su(var)3-9 is

supposedly repressive in these region. However, the fact that HP1a

preferentially binds to the H3K9me-independent gene promoters in these

regions [45] indicates that Su(var)3-9 could have another role in the

pericentromeric regions than only preceding HP1a binding to gene bodies.

Su(var)3-9 is known to interact with the chromo-shadow domain of HP1a,

and maybe this is the primary binding mechanism of Su(var)3-9 in these

particular regions.

Furthermore, SETDB1 and Su(var)3-9 are known to be complementary,

mediating methylation in essentially different regions. However we found, a

Results and discussion

47

bit surprising, that a majority of the genes that are differentially up-or down-

regulated in Su(var)3-9 mutants are correspondingly up- or down-regulated

in a Setdb1 mutant, with an overrepresentation of down-regulated genes.

This indicates that SETDB1 and Su(var)3-9 have more redundant functions,

and they also appear to be more involved in active gene expression than

previously thought. Although, we cannot exclude the possibility that these

observations could be caused by indirect effects, for example by re-direction

of HP1a to normally unbound regions, it could also be that even if different

regions are primary targeted, the same genetic networks will in the end be

affected by the mutants.

In addition, we observed that Su(var)3-9 to our surprise was highly enriched

on chromosome 4, where it has no known function, and when looking at

expression level, we saw that Su(var)3-9 mutants have reduced repression of

chromosome 4 genes. We also noted that mutations in HP1a, Setdb1 or

Su(var)3-9 all individually lead to up-regulation of essentially the same

genes on chromosome 4, meaning that both SETDB1 and Su(var)3-9 appear

to inhibit expression of chromosome 4 genes in an overlapping manner. This

indicates that Su(var)3-9 either has a yet unknown repressive function on

chromosome 4 genes, or this up-regulation is indirectly caused by

redistribution of SETDB1 from the 4th chromosome to compensate for the

absence of HKMTs function in the pericentromeric regions.

HP1a displays a stronger repression of long genes

The repressive effects mediated by HP1a correlate with gene length in a

whole genome comparison. The longer the gene, the more up-regulated it is

in an HP1a mutant, whereas the average HP1a binding per length unit is

unaffected by gene length. This implies that HP1a binds along the gene body

with the same density irrespective of gene length, potentially acting as

“speed-bumps” that slows down the rate of the RNA polymerase II during

the elongation phase of gene transcription. This probably means that the de-

repression caused in an HP1a mutant is stronger for long genes simply

because these genes in total lose more HP1a molecules from the gene body.

HP1a effect in the pericentromeric regions depends on location

We found that the position of the genes can affect level of regulation by

HP1a: in the pericentromeric regions of chromosome arm 2L and 3L, HP1a

binds stronger to genes situated closer to the centromere than to genes

located more distally within the pericentromeric regions. The de-repressing

RESULTS AND DISCUSSION

48

effect in an HP1a mutant in these regions is also stronger the more

proximally located a gene is.

Concluding remarks

It appears that a balance between the amount of euchromatin and

heterochromatin exists in the genome, in which larger heterochromatic

regions have the role of titrating the amount of heterochromatin associated

proteins. If our hypothesis that HP1a has different functions depending on

binding mechanism is correct, it would mean an even more refined mean of

balancing the amount of heterochromatin in a genome, and perhaps also

provide a potentially fast switching mechanism between the two states.

Figure 7. Schematic illustration of the two different types of HP1a binding within

a gene. HP1a binding in the promoter is more stable and leads to activation of gene

expression, potentially by binding in nucleosome core and de-condensing the chromatin

structure. HP1a binding to exons of the gene body depends on H3K9me marks on the histone

tails and represses gene expression.

Results and discussion

49

Conclusions

HP1a represses expression of chromosome 4 active genes, with a

preference for NUEGs; and stimulate expression of pericentromeric

genes, with a slight preference for UEGs.

HP1a binds more predominantly to the promoters of UEGs than the

promoters of NUEGs in general, and on chromosome 4, HP1a

binding is also enhanced at the gene bodies.

HP1a and Su(var)3-9 repress transposon-derived transcripts.

Long genes are more repressed by HP1a, probably because HP1a

binds to gene bodies with the same density, irrespective of gene

length.

Within the pericentromeric regions of chromosome arms 2L and 3L,

HP1a binds to and stimulates expression stronger for genes situated

closer to the heterochromatic centromere.

We propose that HP1a, to better adapt to local chromatin

environment, has evolved opposite functions on gene expression,

and that this is mediated by different binding mechanisms. Either

via a more stable H3K9me-independent binding in the nucleosome

core of promoters which stimulates expression, or a more transient

binding to H3K9 methylation marks at the gene body, which inhibits

gene expression. This gene body binding depends on the presence of

HKMTs (SETDB1 or Su(var)3-9) and on chromosome 4 and it also

depends on the stimulating protein POF.

RESULTS AND DISCUSSION

50

PAPER IV

POF targets roX proximal sites

We made the exciting observation that POF, which displays chromosome 4

specific gene regulatory properties in both sexes, sometimes targets two

specific sites on chromosome X in females, but not in males. These loci were

named PoX1 and PoX2 (POF-on-X) and they contain in total at least seven

genes. The PoX1 site is primarily located in the 3’ end of the Mnt gene and in

a novel non-coding gene located just downstream of Mnt. The PoX2 site

involves the five genes Ck2β, Hsc70-3, CG1578, SelG and CG1840. These

non-4th specific binding sites provide a unique opportunity to study targeting

of POF specifically, but also targeting in general, since the specific locations

of these sites on the X-chromosome provides another very interesting aspect:

both PoX1 and PoX2 are located approximately 200 kb downstream of roX1

and of roX2, respectively. roX1 and roX2 are the two ncRNAs that are part of

the male MSL complex, and also function as high affinity sites for MSL

targeting. The relatively closeness between the roX and the PoX sites thus

provides an additional unique link between these two evolutionary

connected chromosome specific regulatory proteins.

Connection with the MSL complex

We find that POF binding to the PoX sites depends on, but does not require

roX transcription, since a deletion in one or both roX genes results in

decreased, but not entirely absent, PoX-binding. An overexpression of roX2

results in higher PoX targeting frequency. To find out if there are any other

connections between POF binding and the MSL complex binding to these

sites, we compared the POF binding profile in female to a MOF binding

profile in male and find that the MOF binding to the PoX sites is random,

and that no clear connection between POF and MSL complex binding exist

that differs these sites from other genes on the X-chromosome. We could

also not see that these sites differ in expression levels between males and

females, thus the female specific binding to these sites remains a mystery.

Another puzzling aspect is that, although roX transcription appears to

stabilize POF binding to the PoX sites, roX expression in wildtype females is

actually very low or absent, roX1 transcription is only detected in early

embryogenesis and then disappears [230].

Results and discussion

51

Parts of PoX2 functions as POF high affinity target

We used different transgenes or duplications of the PoX genes to investigate

the dependence of local chromatin environment for POF binding, and found

that ectopic autosomal localization of duplications containing either PoX1 or

two of the genes in the PoX2 site, SelG and CG1840, was sufficient to recruit

POF to the site of insertion. In addition, a 6 kb transgene covering SelG and

CG1840 was sufficient to recruit POF targeting.

The binding to the duplicated regions depended on roX transcription in a

similar way as the endogenous sites, whereas the binding to the 6 kb SelG

CG1840 transgene appeared less dependent on roX transcription levels, and

in contrast to the endogenous sites, POF targeted this transgene in both

males and females. We conclude that this 6 kb transgenic region of PoX2

works as a unique POF high-affinity recruitment site, which is independent

on genomic environment. This is a very important discovery, because up to

date, POF binding has been known to be chromosome 4 specific and to

depend on the chromatin environment surrounding the 4th chromosome;

POF cannot bind translocated chromosome 4 regions, nor can POF spread

into segments of other chromosomes translocated onto the tip of

chromosome 4. This is the first described high-affinity site for POF targeting,

and therefore provide a unique opportunity to study the mechanisms behind

POF targeting.

HP1a correlates with POF in the PoX sites

Similar to the situation on the 4th chromosome, HP1a binding co-localizes

with POF also at the PoX sites. In nuclei were no POF was detected at the

PoX sites, no HP1a binding was seen either, indicating that the binding of

HP1a to these sites depends on the presence of POF. Taken together, these

connections between the heterochromatin dependent protein POF, the MSL

complex, and also HP1a, support a model that links dosage compensation

with heterochromatin. The fact that HP1a has been proposed to bind to male

X-chromosome [231], and that in absence of the activating H3S10 kinase

JIL-1 (highly enriched on the male X-chromosome and potentially loosely

attached to the MSL-complex [232,233]), heterochromatic marks such as

H3K9me2 and HP1a are spreading on chromosome arms and most strongly

on chromosome X [234], supports the connection between dosage

compensation and heterochromatin.

RESULTS AND DISCUSSION

52

Conclusions

POF targets the two roX proximal sites, PoX1 and PoX2 on

chromosome X in females, which includes the genes Mnt, Ck2β,

Hsc70-3, CG1578, SelG, CG1840 and the novel, non-coding RNA

RE6469.

Gene expression is not a determinant for the female specific

targeting of POF to PoX1 and PoX2.

POF and HP1a co-localize on the PoX sites.

POF targeting to PoX sites is stabilized by roX activity.

POF targets ectopic PoX1 and partial PoX2 sites inserted on

chromosome 3L in both males and females.

Results and discussion

53

FINAL CONCLUDING REMARKS

Clearly, gene type and chromatin environment have great influence when it

comes to epigenetic regulation, since it appears as the number one

heterochromatin protein, HP1a, depending on context even can adopt

completely opposite functions. HP1a’s functions range from repressing

repetitive regions and potentially deleterious transposable elements, to

upholding expression (maybe by loosening up the chromatin structure at the

promoters) of the ubiquitously expressed UEGs. These are genes important

to maintain housekeeping functions in all cell types, and thus it should be

beneficial with a low energy consuming mechanism which upholds a

constitutive expression of these genes, maybe by a constantly open

chromatin structure. This could be especially important if the genes are

located within a repressive heterochromatic environment, such as the

pericentromeric regions and chromosome 4. HP1a even appears to be able to

alternate between the two opposing functions within the same gene, by

balancing the expression of NUEGs. These are differentially expressed genes

which might require a more fine-tuned expression to maintain correct

expression level in correct tissue.

This can probably also give some clues to the varying buffering capacity we

detected for NUEGs and UEGs, in cases of copy number variation. If, as

indicated by the action of HP1a, NUEGs highly depend on repressive

proteins at the gene body to correctly balance the expression, they should

also be flexible in case an up-regulation is required. UEGs in contrast, whose

actions are more or less constantly required for the maintenance of all cells,

cannot respond as efficiently as NUEGs to copy number reduction. So in

addition to regulating the expression of individual genes, it is also important

for a cell with more general regulatory systems which allows for fast and

simple means of regulating genes with similar functions, i.e. mechanisms

that can enhance or reduce gene expression as a response to different

cellular processes. These types of mechanisms are probably important for

some genes, whereas for other genes (UEGs) it is more essential to maintain

a stable and often high expression level.

ACKNOWLEDGEMENTS

54

ACKNOWLEDGEMENTS

I can honestly say that my years at Molbiol have been the best years of my

life so far! I have learned a lot about science and a lot about myself, and I had

much fun along the way. Mostly thanks to all smart, funny and inspirational

people I have met and come to know over the years, and especially thanks to

the one who personifies all these things, my supervisor Jan Larsson. I feel

truly privileged to have had you as my supervisor! You possess great wisdom,

about science but also about how to shed your knowledge, and simply about

life in general, and I will always be grateful that you saw enough potential in

me to take me on as a student and assigning me these fun projects, and for

supporting and encouraging me throughout these years, even on Monday

mornings . Other people who greatly helped in the formation of this thesis

are Per Stenberg, my almost co-supervisor, if not on paper at least in

practice. Thanks for all ideas and help with interpreting endless numbers of

diagrams and plots, all serious and all less serious discussions (which often

comes in a wonderful mix!), it’s never boring when you are around . And

thanks for introducing me to world of gourmet food (even though I will never

give up on the cheese doodles!). Karin Ekström, the iron lady of the flies,

for all the fly related work, tip on good fly pushing and all fun conversations

in the fika room, and for spreading a really good atmosphere around! I hope

to have your positive attitude, party spirit as well as physique when I retire!

Members of group Jan Larsson: Anna-Mia, “kilen” that holds up the group,

my years in the lab would not have been same without you! Thanks for

always spreading positivism and cheerfulness around you, no matter what,

you have a fantastic ability of making problems and setbacks suddenly

appear easy and trivial! And thanks for all fun conference experiences, late

nights on the dance floors (and tables) etc! Margarida, my partner in

science and in crime! You have brought spice to the group! What have we not

been through together? Conferences, publications, avalanches, awkward

saunas etc etc. Thanks for all our discussions, cooperation, and various

adventures in höga kusten and kittelfjäll, and of course, all beers and movies!

And above all, thanks for being a really good friend! Masha, it so nice that

you joined our group! Thanks for bringing a lot more knowledge and

wisdom, and for boosting dormant projects, Marie-Line, the next-

generation PhD, it’s fun that you have started and I am looking forward to

our up-coming conference trip!

Past members and students of the group: Anders, you are really missed in

our office, your knowledge about basically everything, your sense of humor

ACKNOWLEDGEMENTS

55

and your “bugg” enthusiasm, hope we sometime will get another opportunity

to dazzle the dancefloor! Malin, who was kind enough to leave both her

desk, lab bench and apartment to me when I started in this lab, it’s sad that I

only got to know you for such a short time before you left, it would have been

fun to work more with you, I think we would have gotten along well! Linnéa

and Rafael, for help with the HP1a expression project and the PoX project,

and Denise, really hope to see more of you in the future!

Grp PS: Philge, for your great knowledge in bioinformatics, your useful

scripts and for being kind enough to take the time to explain some of these

things to me as well , the rest of Grp PS: Somyadeep, Aman, Daniel

and Grp YS: Yuri, Tatiana, Sarina, Mischa, Isaac for discussions and

questions on group meetings and/or journal clubs.

The old genetics crew: Anssi, for your kindness and your never-ending

knowledge about genetics. Stefan E, thanks for teaching cooperation and all

the fun “befrielsefester”, but what do you think will become of my defense

party without our no.1 bordercollie? Anna B, thanks for organizing a great

genetics C-course, which really boosted my interest for genetics! And thanks

for movie making experience and “befrielsefester”. Bettan, for wanting to

sneak me in to “rex 27+”, too bad it never happened before the challenge was

gone ;) Also thanks to Magnus for interesting defense parties, and

Marianne R for inspiration!

Other people in the top sphere, the fly-floor: Fredrik, Yasuo, for nice talks,

afterworks, beers and Benjamin Buttons Grp DH: Dan, Jesper, Jens-

Ola, Inez, Martin, Sajna, Hairu and Silvia for creating a very nice

homey feeling in the fika room! And also great thanks to all other people on

this floor: Åsa, Ruth, Caroline, Camilla, Ingrid, Erik, John, Linn,

Sa, Murat, Barbara, Mahsa, Jesper, Behzad, Sofie and all past fly

floor/UCMP people; Anna L, Anna-Karin, Karin, Jana, Ulrika,

Dimitri, Margret, Mazen, Magdalena, Gaurav, Mattias, Therese N,

Olga, Maria W, Malin L-P, Anders O, and many more, for nice fika

room chats, for brightening up the corridors and for creating a genuine nice

and creative environment!

Friends from Molbiol (present or out on their own new adventures):

Therese E, I am so sad that you have left umeå, we have had a lot of fun

together! Thanks for all craft nights, you are a great source of inspiration,

and thanks for being a genuine good friend! Luckily, Sundsvall is not that far

away, and hope we will have more vacations together on Crete in the future!

ACKNOWLEDGEMENTS

56

With Erik N, Gunnar and Karin! Erik T, you are one of a kind, thanks for

being you, my years at molbiol would have been much less fun without you,

you’re always up for stuff and ready for adventures, your always kind and

funny, and take things in the right way! And thanks for your endless patients

in teaching beach-volleyboll! You are welcome to come and sink my canoe

again at any time you want in the future . Linn J, thanks for all fun events,

laughs, craft nights, spex movie making etc etc, if you ever resign from

science I think you should become an actress! John S-H, you are really fun

to be around, thanks for all lunches, fika, movie making etc, and thanks for

doing a great job as our fika over-lord, it would not have worked so well

without you! Chaz and Linus, I really miss you guys! You can light up every

occasion, and Chaz thanks for endless craft nights and for having the same

enthusiasm for costume parties as me! You should move back to umeå and

open “the drunken bead”, but maybe in the future our paths will cross again,

who knows Hande, thanks for all good times and nice discussions we

have had! Sa, for many pleasant dinners, craft nights and movie planning

nights in your home, you deserve the best! David G, probably the funniest

person I have met, it’s too bad you left umeå so early (in my perspective at

least) hope you survive the trip from Göteborg back to Umeå and my defense

party without any major mishaps ;) Sara R, for good times at the lunch

table, parties, spontaneous bonfires at parties ;) and for being an excellent

spex movie planner! Christina S, for crafts nights and BBQ at your balcony,

hope you are have a good time in Scotland! Anne-Laure, thanks for

movies, adventure on höga kusten and afterworks. The members of the lunch

and fika cluster (John, Linn, Erik, Margarida, Marie-Line, Caroline,

Christopher, Isabelle, Sajna, Victoria, Lisa, Jessica), thanks to all of you for

all great lunch discussions about small and big things, the lunches with you

guys can really make my day! Caroline, thanks for all nice chats and good

luck with the new family! Christopher, you have a talent for balancing

humor, bitterness and irony in a lovely mixture. Isabelle, for nice advices

about life, dissertations etc. Sajna, for all your sharp and funny stories.

Victoria, for teaching us pottery. Lisa, for lightening up my days by teasing

Andersson Jessica, our coffee-junkie, I think you’ve been gone too long!

Hasan, Tony, and previous presidents of Beer corner for maintaining a

good selection of beers! Hao, Hasan, Kristoffer L, Anais, Ayad,

Patrik, for nice teaching cooperation. Marek, for fixing and installing

computers, and Mediafolket, for always being helpful and nice! And thanks

to the fun people at beer corner, the molbiol crafting gang, and all other

people at the molbiol department that I have not mentioned but not

forgotten!

ACKNOWLEDGEMENTS

57

Människorna ute i den andra världen: Emma, det närmsta en syster jag har

kommit, kan inte minnas nån tid jag inte känt dig och jag tror mitt liv hade

varit väldigt annorlunda om jag inte hade känt dig, tack för allt vi har upplevt

ihop (det är omöjligt att lista upp) och för att du alltid har varit min vän!

Mogge, tack för att du tar hand om emma ;) Du är en hemskt roligt person

och jag är glad att ni har gift er! Vera, tack för att du varit min vän i alla

dessa år, jag saknar dig! Det är himla tur att det finns telefon och billiga

abonnemang! Hoppas vi nån dag kommer bo närmre varandra! Maria,

utan dig hade jag nog haft grava uppdämda psykologiska problem ;) tack för

din fina vänskap och för alla kloka råd och tankar om allt! Mats, ni två

passar väldigt bra ihop och jag ser fram emot att få lära känna dig ännu

bättre! Monika, tack för året som kombos och allt roligt vi har haft efter det,

jag hoppas på mycket mer i en snar framtid! Jag och umeå har saknat dig,

även om du inte saknar umeå ;) och tack för att du betrodde mig med dina

mysiga katter!

Slutligen min familj, den bästa man skulle kunna fått: Det är väldigt svårt att

beskriva i ord vad ni betyder för mig, Mamma och Pappa, tack för livets

gåva och all stöttning ni ger mig, för att ni alltid finns där! Mamma, jag

hoppas jag blir som du, jag tror inte det finns nån klokare kvinna, Pappa, för

att jag fick ärva ditt intresse för naturvetenskap, och tack för att du alltid,

alltid ställt upp och funnit till hands när jag behövt! Från att lösa

matteproblem och pussla ihop vattenlås till att köra 100 mil för att hjälpa

mig flytta, Arvid och Markus, mina älskade mallis, småbröder, vad vore väl

nånting utan er? Markus, minns mig när du blir rik och berömd

spelutvecklare Arvid, jag saknar att ha dig i närheten, varför måste ni bo

så långt bort! Mary, välkommen in i familjen, hoppas verkligen vi får fler

resor till solen tillsammas, eller åtminstone fler mysdagar i stugan så länge

Farfar, som tydligt visat att det är annat virke i den generationen och att

man kan klara det mesta på egen hand, bara envisheten och viljan finns där,

med eller utan brutna lårbenshalsar.

“Nya möjligheter är ofta förklädda till smärtsamma slut.” ~ Lao Zi

REFERENCES

58

REFERENCES

1. Khorasanizadeh S (2004) The nucleosome: from genomic organization to

genomic regulation. Cell 116: 259-272. 2. Olins AL, Olins DE (1974) Spheroid chromatin units (v bodies). Science

183: 330-332. 3. Oudet P, Gross-Bellard M, Chambon P (1975) Electron microscopic and

biochemical evidence that chromatin structure is a repeating unit. Cell 4: 281-300.

4. Thomas JO, Kornberg RD (1975) An octamer of histones in chromatin and free in solution. Proc Natl Acad Sci U S A 72: 2626-2630.

5. Luger K, Mäder AW, Richmond RK, Sargent DF, Richmond TJ (1997) Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature 389: 251-260.

6. Happel N, Doenecke D (2009) Histone H1 and its isoforms: contribution to chromatin structure and function. Gene 431: 1-12.

7. Kornberg RD (1974) Chromatin structure: a repeating unit of histones and DNA. Science 184: 868-871.

8. Wong H, Victor JM, Mozziconacci J (2007) An all-atom model of the chromatin fiber containing linker histones reveals a versatile structure tuned by the nucleosomal repeat length. PLoS One 2: e877.

9. Robinson PJ, Fairall L, Huynh VA, Rhodes D (2006) EM measurements define the dimensions of the "30-nm" chromatin fiber: evidence for a compact, interdigitated structure. Proc Natl Acad Sci U S A 103: 6506-6511.

10. Zheng C, Hayes JJ (2003) Structures and interactions of the core histone tail domains. Biopolymers 68: 539-546.

11. Gräff J, Tsai LH (2013) Histone acetylation: molecular mnemonics on the chromatin. Nat Rev Neurosci 14: 97-111.

12. Smith CM, Gafken PR, Zhang Z, Gottschling DE, Smith JB, et al. (2003) Mass spectrometric quantification of acetylation at specific lysines within the amino-terminal tail of histone H4. Anal Biochem 316: 23-33.

13. Shogren-Knaak M, Ishii H, Sun JM, Pazin MJ, Davie JR, et al. (2006) Histone H4-K16 acetylation controls chromatin structure and protein interactions. Science 311: 844-847.

14. Tse C, Sera T, Wolffe AP, Hansen JC (1998) Disruption of higher-order folding by core histone acetylation dramatically enhances transcription of nucleosomal arrays by RNA polymerase III. Mol Cell Biol 18: 4629-4638.

15. Lyko F, Ramsahoye BH, Jaenisch R (2000) DNA methylation in Drosophila melanogaster. Nature 408: 538-540.

REFERENCES

59

16. Peters AH, Kubicek S, Mechtler K, O'Sullivan RJ, Derijck AA, et al. (2003) Partitioning and plasticity of repressive histone methylation states in mammalian chromatin. Mol Cell 12: 1577-1589.

17. Santos-Rosa H, Schneider R, Bannister AJ, Sherriff J, Bernstein BE, et al. (2002) Active genes are tri-methylated at K4 of histone H3. Nature 419: 407-411.

18. Mikkelsen TS, Ku M, Jaffe DB, Issac B, Lieberman E, et al. (2007) Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature 448: 553-560.

19. Martin C, Zhang Y (2005) The diverse functions of histone lysine methylation. Nat Rev Mol Cell Biol 6: 838-849.

20. Ebert A, Schotta G, Lein S, Kubicek S, Krauss V, et al. (2004) Su(var) genes regulate the balance between euchromatin and heterochromatin in Drosophila. Genes Dev 18: 2973-2983.

21. Grewal SI, Elgin SC (2002) Heterochromatin: new possibilities for the inheritance of structure. Curr Opin Genet Dev 12: 178-187.

22. Eissenberg JC, Reuter G (2009) Cellular mechanism for targeting heterochromatin formation in Drosophila. Int Rev Cell Mol Biol 273: 1-47.

23. Richards EJ, Elgin SC (2002) Epigenetic codes for heterochromatin formation and silencing: rounding up the usual suspects. Cell 108: 489-500.

24. Gowen JW, Gay EH (1933) Effect of temperature on eversporting eye color in Drosophila melanogaster. Science 77: 312.

25. Dimitri P, Pisano C (1989) Position effect variegation in Drosophila melanogaster: relationship between suppression effect and the amount of Y chromosome. Genetics 122: 793-800.

26. Gowen JW, Gay EH (1934) Chromosome constitution and behavior in eversporting and mottling in Drosophila melanogaster. Genetics 19: 189-208.

27. James TC, Eissenberg JC, Craig C, Dietrich V, Hobson A, et al. (1989) Distribution patterns of HP1, a heterochromatin-associated nonhistone chromosomal protein of Drosophila. Eur J Cell Biol 50: 170-180.

28. Ebert A, Lein S, Schotta G, Reuter G (2006) Histone modification and the control of heterochromatic gene silencing in Drosophila. Chromosome Res 14: 377-392.

29. Carvalho A, Koerich L, Clark A (2009) Origin and evolution of Y chromosomes: Drosophila tales. Trends Genet: 270-277.

30. Filion GJ, van Bemmel JG, Braunschweig U, Talhout W, Kind J, et al. (2010) Systematic protein location mapping reveals five principal chromatin types in Drosophila cells. Cell 143: 212-224.

REFERENCES

60

31. Zhang P, Du J, Sun B, Dong X, Xu G, et al. (2006) Structure of human MRG15 chromo domain and its binding to Lys36-methylated histone H3. Nucleic Acids Res 34: 6621-6628.

32. Kharchenko PV, Alekseyenko AA, Schwartz YB, Minoda A, Riddle NC, et al. (2011) Comprehensive analysis of the chromatin landscape in Drosophila melanogaster. Nature 471: 480-485.

33. Wallrath LL (1998) Unfolding the mysteries of heterochromatin. Curr Opin Genet Dev 8: 147-153.

34. Eissenberg JC, James TC, Foster-Hartnett DM, Hartnett T, Ngan V, et al. (1990) Mutation in a heterochromatin-specific chromosomal protein is associated with suppression of position-effect variegation in Drosophila melanogaster. Proc Natl Acad Sci U S A 87: 9923-9927.

35. Bannister AJ, Zegerman P, Partridge JF, Miska EA, Thomas JO, et al. (2001) Selective recognition of methylated lysine 9 on histone H3 by the HP1 chromo domain. Nature 410: 120-124.

36. Lachner M, O'Carroll D, Rea S, Mechtler K, Jenuwein T (2001) Methylation of histone H3 lysine 9 creates a binding site for HP1 proteins. Nature 410: 116-120.

37. Vermaak D, Malik HS (2009) Multiple roles for heterochromatin protein 1 genes in Drosophila. Annu Rev Genet 43: 467-492.

38. Aasland R, Stewart AF (1995) The chromo shadow domain, a second chromo domain in heterochromatin-binding protein 1, HP1. Nucleic Acids Res 23: 3168-3173.

39. Cowieson NP, Partridge JF, Allshire RC, McLaughlin PJ (2000) Dimerisation of a chromo shadow domain and distinctions from the chromodomain as revealed by structural analysis. Curr Biol 10: 517-525.

40. Mishima Y, Watanabe M, Kawakami T, Jayasinghe CD, Otani J, et al. (2013) Hinge and chromoshadow of HP1α participate in recognition of K9 methylated histone H3 in nucleosomes. J Mol Biol 425: 54-70.

41. Fanti L, Berloco M, Piacentini L, Pimpinelli S (2003) Chromosomal distribution of heterochromatin protein 1 (HP1) in Drosophila: a cytological map of euchromatic HP1 binding sites. Genetica 117: 135-147.

42. Cryderman DE, Grade SK, Li Y, Fanti L, Pimpinelli S, et al. (2005) Role of Drosophila HP1 in euchromatic gene expression. Dev Dyn 232: 767-774.

43. Riddle NC, Minoda A, Kharchenko PV, Alekseyenko AA, Schwartz YB, et al. (2011) Plasticity in patterns of histone modifications and chromosomal proteins in Drosophila heterochromatin. Genome Res 21: 147-163.

44. Yin H, Sweeney S, Raha D, Snyder M, Lin H (2011) A high-resolution whole-genome map of key chromatin modifications in the adult Drosophila melanogaster. PLoS Genet 7: e1002380.

REFERENCES

61

45. Figueiredo MLA, Philip P, Stenberg P, Larsson J (2012) HP1a recruitment to promoters is independent of H3K9 methylation in Drosophila melanogaster. PLoS Genet 8: e1003061.

46. Johansson AM, Stenberg P, Pettersson F, Larsson J (2007) POF and HP1 bind expressed exons, suggesting a balancing mechanism for gene regulation. PLoS Genet 3: e209.

47. de Wit E, Greil F, van Steensel B (2007) High-resolution mapping reveals links of HP1 with active and inactive chromatin components. PLoS Genet 3: e38.

48. Johansson AM, Stenberg P, Bernhardsson C, Larsson J (2007) Painting of fourth and chromosome-wide regulation of the 4th chromosome in Drosophila melanogaster. EMBO J 26: 2307-2316.

49. Liu LP, Ni JQ, Shi YD, Oakeley EJ, Sun FL (2005) Sex-specific role of Drosophila melanogaster HP1 in regulating chromatin structure and gene transcription. Nat Genet 37: 1361-1366.

50. De Lucia F, Ni JQ, Vaillant C, Sun FL (2005) HP1 modulates the transcription of cell-cycle regulators in Drosophila melanogaster. Nucleic Acids Res 33: 2852-2858.

51. Schwaiger M, Kohler H, Oakeley EJ, Stadler MB, Schübeler D (2010) Heterochromatin protein 1 (HP1) modulates replication timing of the Drosophila genome. Genome Res 20: 771-780.

52. Piacentini L, Fanti L, Berloco M, Perrini B, Pimpinelli S (2003) Heterochromatin protein 1 (HP1) is associated with induced gene expression in Drosophila euchromatin. J Cell Biol 161: 707-714.

53. Piacentini L, Fanti L, Negri R, Del Vescovo V, Fatica A, et al. (2009) Heterochromatin protein 1 (HP1a) positively regulates euchromatic gene expression through RNA transcript association and interaction with hnRNPs in Drosophila. PLoS Genet 5: e1000670.

54. Riddle NC, Jung YL, Gu T, Alekseyenko AA, Asker D, et al. (2012) Enrichment of HP1a on Drosophila chromosome 4 genes creates an alternate chromatin structure critical for regulation in this heterochromatic domain. PLoS Genet 8: e1002954.

55. Hearn MG, Hedrick A, Grigliatti TA, Wakimoto BT (1991) The effect of modifiers of position-effect variegation on the variegation of heterochromatic genes of Drosophila melanogaster. Genetics 128: 785-797.

56. Wakimoto BT, Hearn MG (1990) The effects of chromosome rearrangements on the expression of heterochromatic genes in chromosome 2L of Drosophila melanogaster. Genetics 125: 141-154.

57. Clegg NJ, Honda BM, Whitehead IP, Grigliatti TA, Wakimoto B, et al. (1998) Suppressors of position-effect variegation in Drosophila melanogaster affect expression of the heterochromatic gene light in the absence of a chromosome rearrangement. Genome 41: 495-503.

REFERENCES

62

58. Lu BY, Emtage PC, Duyf BJ, Hilliker AJ, Eissenberg JC (2000) Heterochromatin protein 1 is required for the normal expression of two heterochromatin genes in Drosophila. Genetics 155: 699-708.

59. Schulze SR, Sinclair DA, Fitzpatrick KA, Honda BM (2005) A genetic and molecular characterization of two proximal heterochromatic genes on chromosome 3 of Drosophila melanogaster. Genetics 169: 2165-2177.

60. Yasuhara JC, Wakimoto BT (2006) Oxymoron no more: the expanding world of heterochromatic genes. Trends Genet 22: 330-338.

61. Smothers JF, Henikoff S (2001) The hinge and chromo shadow domain impart distinct targeting of HP1-like proteins. Mol Cell Biol 21: 2555-2569.

62. Font-Burgada J, Rossell D, Auer H, Azorín F (2008) Drosophila HP1c isoform interacts with the zinc-finger proteins WOC and Relative-of-WOC to regulate gene expression. Genes Dev 22: 3007-3023.

63. Vermaak D, Henikoff S, Malik HS (2005) Positive selection drives the evolution of rhino, a member of the heterochromatin protein 1 family in Drosophila. PLoS Genet 1: 96-108.

64. Mis J, Ner SS, Grigliatti TA (2006) Identification of three histone methyltransferases in Drosophila: dG9a is a suppressor of PEV and is required for gene silencing. Mol Genet Genomics 275: 513-526.

65. Seum C, Bontron S, Reo E, Delattre M, Spierer P (2007) Drosophila G9a is a nonessential gene. Genetics 177: 1955-1957.

66. Stabell M, Eskeland R, Bjørkmo M, Larsson J, Aalen RB, et al. (2006) The Drosophila G9a gene encodes a multi-catalytic histone methyltransferase required for normal development. Nucleic Acids Res 34: 4609-4621.

67. Brower-Toland B, Riddle NC, Jiang H, Huisinga KL, Elgin SC (2009) Multiple SET methyltransferases are required to maintain normal heterochromatin domains in the genome of Drosophila melanogaster. Genetics 181: 1303-1319.

68. Rea S, Eisenhaber F, O'Carroll D, Strahl BD, Sun ZW, et al. (2000) Regulation of chromatin structure by site-specific histone H3 methyltransferases. Nature 406: 593-599.

69. Czermin B, Schotta G, Hülsmann BB, Brehm A, Becker PB, et al. (2001) Physical and functional association of SU(VAR)3-9 and HDAC1 in Drosophila. EMBO Rep 2: 915-919.

70. Schotta G, Ebert A, Krauss V, Fischer A, Hoffmann J, et al. (2002) Central role of Drosophila SU(VAR)3-9 in histone H3-K9 methylation and heterochromatic gene silencing. EMBO J 21: 1121-1131.

71. Tschiersch B, Hofmann A, Krauss V, Dorn R, Korge G, et al. (1994) The protein encoded by the Drosophila position-effect variegation

REFERENCES

63

suppressor gene Su(var)3-9 combines domains of antagonistic regulators of homeotic gene complexes. EMBO J 13: 3822-3831.

72. Seum C, Reo E, Peng H, Rauscher FJ, Spierer P, et al. (2007) Drosophila SETDB1 is required for chromosome 4 silencing. PLoS Genet 3: e76.

73. Tzeng TY, Lee CH, Chan LW, Shen CK (2007) Epigenetic regulation of the Drosophila chromosome 4 by the histone H3K9 methyltransferase dSETDB1. Proc Natl Acad Sci U S A 104: 12691-12696.

74. Clough E, Moon W, Wang S, Smith K, Hazelrigg T (2007) Histone methylation is required for oogenesis in Drosophila. Development 134: 157-165.

75. Stabell M, Bjørkmo M, Aalen RB, Lambertsson A (2006) The Drosophila SET domain encoding gene dEset is essential for proper development. Hereditas 143: 177-188.

76. Yoon J, Lee KS, Park JS, Yu K, Paik SG, et al. (2008) dSETDB1 and SU(VAR)3-9 sequentially function during germline-stem cell differentiation in Drosophila melanogaster. PLoS One 3: e2234.

77. Koch CM, Honemann-Capito M, Egger-Adam D, Wodarz A (2009) Windei, the Drosophila homolog of mAM/MCAF1, is an essential cofactor of the H3K9 methyl transferase dSETDB1/Eggless in germ line development. PLoS Genet 5: e1000644.

78. Nakayama J, Rice JC, Strahl BD, Allis CD, Grewal SI (2001) Role of histone H3 lysine 9 methylation in epigenetic control of heterochromatin assembly. Science 292: 110-113.

79. Jacobs SA, Khorasanizadeh S (2002) Structure of HP1 chromodomain bound to a lysine 9-methylated histone H3 tail. Science 295: 2080-2083.

80. Jacobs SA, Taverna SD, Zhang Y, Briggs SD, Li J, et al. (2001) Specificity of the HP1 chromo domain for the methylated N-terminus of histone H3. EMBO J 20: 5232-5241.

81. Mottus R, Sobel RE, Grigliatti TA (2000) Mutational analysis of a histone deacetylase in Drosophila melanogaster: missense mutations suppress gene silencing associated with position effect variegation. Genetics 154: 657-668.

82. Slotkin RK, Martienssen R (2007) Transposable elements and the epigenetic regulation of the genome. Nat Rev Genet 8: 272-285.

83. Huang XA, Yin H, Sweeney S, Raha D, Snyder M, et al. (2013) A major epigenetic programming mechanism guided by piRNAs. Dev Cell 24: 502-516.

84. McClintock B (1950) The origin and behavior of mutable loci in maize. Proc Natl Acad Sci U S A 36: 344-355.

85. Craig NL (1995) Unity in transposition reactions. Science 270: 253-254.

REFERENCES

64

86. Rao M, Sockanathan S (2005) Molecular mechanisms of RNAi: implications for development and disease. Birth Defects Res C Embryo Today 75: 28-42.

87. Fanti L, Dorer DR, Berloco M, Henikoff S, Pimpinelli S (1998) Heterochromatin protein 1 binds transgene arrays. Chromosoma 107: 286-292.

88. Greil F, van der Kraan I, Delrow J, Smothers JF, de Wit E, et al. (2003) Distinct HP1 and Su(var)3-9 complexes bind to sets of developmentally coexpressed genes depending on chromosomal location. Genes Dev 17: 2825-2838.

89. Smith CD, Shu S, Mungall CJ, Karpen GH (2007) The Release 5.1 annotation of Drosophila melanogaster heterochromatin. Science 316: 1586-1591.

90. Adams MD, Celniker SE, Holt RA, Evans CA, Gocayne JD, et al. (2000) The genome sequence of Drosophila melanogaster. Science 287: 2185-2195.

91. Reiter LT, Potocki L, Chien S, Gribskov M, Bier E (2001) A systematic analysis of human disease-associated gene sequences in Drosophila melanogaster. Genome Res 11: 1114-1125.

92. Skaletsky H, Kuroda-Kawaguchi T, Minx PJ, Cordum HS, Hillier L, et al. (2003) The male-specific region of the human Y chromosome is a mosaic of discrete sequence classes. Nature 423: 825-837.

93. Koopman P, Gubbay J, Vivian N, Goodfellow P, Lovell-Badge R (1991) Male development of chromosomally female mice transgenic for Sry. Nature 351: 117-121.

94. Carvalho AB (2002) Origin and evolution of the Drosophila Y chromosome. Curr Opin Genet Dev 12: 664-668.

95. Lucchesi JC, Kelly WG, Panning B (2005) Chromatin remodeling in dosage compensation. Annu Rev Genet 39: 615-651.

96. Lyon MF (1961) Gene action in the X-chromosome of the mouse (Mus musculus L.). Nature 190: 372-373.

97. Barr ML, Carr DH (1962) Correlations between sex chromatin and sex chromosomes. Acta Cytol 6: 34-45.

98. Lee JT, Strauss WM, Dausman JA, Jaenisch R (1996) A 450 kb transgene displays properties of the mammalian X-inactivation center. Cell 86: 83-94.

99. Clemson CM, McNeil JA, Willard HF, Lawrence JB (1996) XIST RNA paints the inactive X chromosome at interphase: evidence for a novel RNA involved in nuclear/chromosome structure. J Cell Biol 132: 259-275.

100. Herzing LB, Romer JT, Horn JM, Ashworth A (1997) Xist has properties of the X-chromosome inactivation centre. Nature 386: 272-275.

REFERENCES

65

101. Zhao J, Sun BK, Erwin JA, Song JJ, Lee JT (2008) Polycomb proteins targeted by a short repeat RNA to the mouse X chromosome. Science 322: 750-756.

102. Heard E, Disteche CM (2006) Dosage compensation in mammals: fine-tuning the expression of the X chromosome. Genes Dev 20: 1848-1867.

103. Chadwick BP, Willard HF (2003) Chromatin of the Barr body: histone and non-histone proteins associated with or excluded from the inactive X chromosome. Hum Mol Genet 12: 2167-2178.

104. Deng X, Hiatt JB, Nguyen DK, Ercan S, Sturgill D, et al. (2011) Evidence for compensatory upregulation of expressed X-linked genes in mammals, Caenorhabditis elegans and Drosophila melanogaster. Nat Genet 43: 1043‐1047.

105. Gupta V, Parisi M, Sturgill D, Nuttall R, Doctolero M, et al. (2006) Global analysis of X-chromosome dosage compensation. J Biol 5: 3.

106. Nguyen DK, Disteche CM (2006) Dosage compensation of the active X chromosome in mammals. Nat Genet 38: 47-53.

107. Adler DA, Rugarli EI, Lingenfelter PA, Tsuchiya K, Poslinski D, et al. (1997) Evidence of evolutionary up-regulation of the single active X chromosome in mammals based on Clc4 expression levels in Mus spretus and Mus musculus. Proc Natl Acad Sci U S A 94: 9244-9248.

108. Prestel M, Feller C, Becker PB (2010) Dosage compensation and the global re-balancing of aneuploid genomes. Genome Biol 11: 216.

109. Vallot C, Huret C, Lesecque Y, Resch A, Oudrhiri N, et al. (2013) XACT, a long noncoding transcript coating the active X chromosome in human pluripotent cells. Nat Genet 45: 239-241.

110. Xiong Y, Chen X, Chen Z, Wang X, Shi S, et al. (2010) RNA sequencing shows no dosage compensation of the active X-chromosome. Nat Genet 42: 1043-1047.

111. Smith ER, Pannuti A, Gu W, Steurnagel A, Cook RG, et al. (2000) The drosophila MSL complex acetylates histone H4 at lysine 16, a chromatin modification linked to dosage compensation. Mol Cell Biol 20: 312-318.

112. Kadlec J, Hallacli E, Lipp M, Holz H, Sanchez-Weatherby J, et al. (2011) Structural basis for MOF and MSL3 recruitment into the dosage compensation complex by MSL1. Nat Struct Mol Biol 18: 142-149.

113. Chang KA, Kuroda MI (1998) Modulation of MSL1 abundance in female Drosophila contributes to the sex specificity of dosage compensation. Genetics 150: 699-709.

114. Morales V, Straub T, Neumann MF, Mengus G, Akhtar A, et al. (2004) Functional integration of the histone acetyltransferase MOF into the dosage compensation complex. EMBO J 23: 2258-2268.

115. Copps K, Richman R, Lyman LM, Chang KA, Rampersad-Ammons J, et al. (1998) Complex formation by the Drosophila MSL proteins: role

REFERENCES

66

of the MSL2 RING finger in protein complex assembly. EMBO J 17: 5409-5417.

116. Scott MJ, Pan LL, Cleland SB, Knox AL, Heinrich J (2000) MSL1 plays a central role in assembly of the MSL complex, essential for dosage compensation in Drosophila. EMBO J 19: 144-155.

117. Hallacli E, Lipp M, Georgiev P, Spielman C, Cusack S, et al. (2012) Msl1-mediated dimerization of the dosage compensation complex is essential for male X-chromosome regulation in Drosophila. Mol Cell 48: 587-600.

118. Bashaw GJ, Baker BS (1995) The msl-2 dosage compensation gene of Drosophila encodes a putative DNA-binding protein whose expression is sex specifically regulated by Sex-lethal. Development 121: 3245-3258.

119. Zhou S, Yang Y, Scott MJ, Pannuti A, Fehr KC, et al. (1995) Male-specific lethal 2, a dosage compensation gene of Drosophila, undergoes sex-specific regulation and encodes a protein with a RING finger and a metallothionein-like cysteine cluster. EMBO J 14: 2884-2895.

120. Kelley RL, Wang J, Bell L, Kuroda MI (1997) Sex lethal controls dosage compensation in Drosophila by a non-splicing mechanism. Nature 387: 195-199.

121. Kelley RL, Solovyeva I, Lyman LM, Richman R, Solovyev V, et al. (1995) Expression of msl-2 causes assembly of dosage compensation regulators on the X chromosomes and female lethality in Drosophila. Cell 81: 867-877.

122. Bashaw GJ, Baker BS (1997) The regulation of the Drosophila msl-2 gene reveals a function for Sex-lethal in translational control. Cell 89: 789-798.

123. Johansson AM, Allgardsson A, Stenberg P, Larsson J (2011) msl2 mRNA is bound by free nuclear MSL complex in Drosophila melanogaster. Nucleic Acids Res 39: 6428-6439.

124. Koonin EV, Zhou S, Lucchesi JC (1995) The chromo superfamily: new members, duplication of the chromo domain and possible role in delivering transcription regulators to chromatin. Nucleic Acids Res 23: 4229-4233.

125. Buscaino A, Köcher T, Kind JH, Holz H, Taipale M, et al. (2003) MOF-regulated acetylation of MSL-3 in the Drosophila dosage compensation complex. Mol Cell 11: 1265-1277.

126. Lee CG, Chang KA, Kuroda MI, Hurwitz J (1997) The NTPase/helicase activities of Drosophila maleless, an essential factor in dosage compensation. EMBO J 16: 2671-2681.

127. Wang CI, Alekseyenko AA, Leroy G, Elia AE, Gorchakov AA, et al. (2013) Chromatin proteins captured by ChIP-mass spectrometry are linked

REFERENCES

67

to dosage compensation in Drosophila. Nat Struct Mol Biol 20: 202-209.

128. Richter L, Bone JR, Kuroda MI (1996) RNA-dependent association of the Drosophila maleless protein with the male X chromosome. Genes to Cells 1: 325-336.

129. Morra R, Yokoyama R, Ling H, Lucchesi JC (2011) Role of the ATPase/helicase maleless (MLE) in the assembly, targeting, spreading and function of the male-specific lethal (MSL) complex of Drosophila. Epigenetics Chromatin 4: 6.

130. Oh H, Park Y, Kuroda MI (2003) Local spreading of MSL complexes from roX genes on the Drosophila X chromosome. Genes Dev 17: 1334-1339.

131. Meller VH, Gordadze PR, Park Y, Chu X, Stuckenholz C, et al. (2000) Ordered assembly of roX RNAs into MSL complexes on the dosage-compensated X chromosome in Drosophila. Curr Biol 10: 136-143.

132. Morra R, Smith ER, Yokoyama R, Lucchesi JC (2008) The MLE subunit of the Drosophila MSL complex uses its ATPase activity for dosage compensation and its helicase activity for targeting. Mol Cell Biol 28: 958-966.

133. Kotlikova IV, Demakova OV, Semeshin VF, Shloma VV, Boldyreva LV, et al. (2006) The Drosophila dosage compensation complex binds to polytene chromosomes independently of developmental changes in transcription. Genetics 172: 963-974.

134. Stuckenholz C, Kageyama Y, Kuroda MI (1999) Guilt by association, non-coding RNAs, chromosome-specific proteins and dosage compensation in Drosophila. Trends Genet 15: 454-458.

135. Akhtar A, Becker PB (2000) Activation of transcription through histone H4 acetylation by MOF, an acetyltransferase essential for dosage compensation in Drosophila. Mol Cell 5: 367-375.

136. Hilfiker A, Hilfiker-Kleiner D, Pannuti A, Lucchesi JC (1997) mof, a putative acetyl transferase gene related to the Tip60 and MOZ human genes and to the SAS genes of yeast, is required for dosage compensation in Drosophila. EMBO J 16: 2054-2060.

137. Kind J, Vaquerizas JM, Gebhardt P, Gentzel M, Luscombe NM, et al. (2008) Genome-wide analysis reveals MOF as a key regulator of dosage compensation and gene expression in Drosophila. Cell 133: 813-828.

138. Bhadra U, Pal-Bhadra M, Birchler JA (1999) Role of the male specific lethal (msl) genes in modifying the effects of sex chromosomal dosage in Drosophila. Genetics 152: 249-268.

139. Lam KC, Mühlpfordt F, Vaquerizas JM, Raja SJ, Holz H, et al. (2012) The NSL complex regulates housekeeping genes in Drosophila. PLoS Genet 8: e1002736.

REFERENCES

68

140. Raja SJ, Charapitsa I, Conrad T, Vaquerizas JM, Gebhardt P, et al. (2010) The nonspecific lethal complex is a transcriptional regulator in Drosophila. Mol Cell 38: 827-841.

141. Prestel M, Feller C, Straub T, Mitlöhner H, Becker PB (2010) The activation potential of MOF is constrained for dosage compensation. Mol Cell 38: 815-826.

142. Franke A, Baker BS (1999) The rox1 and rox2 RNAs are essential components of the Compensasome, which mediates dosage compensation in Drosophila. Mol Cell 4: 117-122.

143. Meller VH, Wu KH, Roman G, Kuroda MI, Davis RL (1997) roX1 RNA paints the X chromosome of male Drosophila and is regulated by the dosage compensation system. Cell 88: 445-457.

144. Meller VH, Rattner BP (2002) The roX genes encode redundant male-specific lethal transcripts required for targeting of the MSL complex. EMBO J 21: 1084-1091.

145. Kelley RL, Meller VH, Gordadze PR, Roman G, Davis RL, et al. (1999) Epigenetic spreading of the Drosophila dosage compensation complex from roX RNA genes into flanking chromatin. Cell 98: 513-522.

146. Larschan E, Alekseyenko AA, Gortchakov AA, Peng S, Li B, et al. (2007) MSL complex is attracted to genes marked by H3K36 trimethylation using a sequence-independent mechanism. Mol Cell 28: 121-133.

147. Straub T, Zabel A, Gilfillan GD, Feller C, Becker PB (2013) Different chromatin interfaces of the Drosophila dosage compensation complex revealed by high-shear ChIP-seq. Genome Res 23: 473–485.

148. Alekseyenko AA, Peng S, Larschan E, Gorchakov AA, Lee OK, et al. (2008) A sequence motif within chromatin entry sites directs MSL establishment on the Drosophila X chromosome. Cell 134: 599-609.

149. Palmer MJ, Richman R, Richter L, Kuroda MI (1994) Sex-specific regulation of the male-specific lethal-1 dosage compensation gene in Drosophila. Genes Dev 8: 698-706.

150. Demakova OV, Kotlikova IV, Gordadze PR, Alekseyenko AA, Kuroda MI, et al. (2003) The MSL complex levels are critical for its correct targeting to the chromosomes in Drosophila melanogaster. Chromosoma 112: 103-115.

151. Lyman LM, Copps K, Rastelli L, Kelley RL, Kuroda MI (1997) Drosophila male-specific lethal-2 protein: structure/function analysis and dependence on MSL-1 for chromosome association. Genetics 147: 1743-1753.

152. Alekseyenko AA, Larschan E, Lai WR, Park PJ, Kuroda MI (2006) High-resolution ChIP-chip analysis reveals that the Drosophila MSL complex selectively identifies active genes on the male X chromosome. Genes Dev 20: 848-857.

REFERENCES

69

153. Gilfillan GD, Straub T, de Wit E, Greil F, Lamm R, et al. (2006) Chromosome-wide gene-specific targeting of the Drosophila dosage compensation complex. Genes Dev 20: 858-870.

154. Sural TH, Peng S, Li B, Workman JL, Park PJ, et al. (2008) The MSL3 chromodomain directs a key targeting step for dosage compensation of the Drosophila melanogaster X chromosome. Nat Struct Mol Biol 15: 1318-1325.

155. Bell O, Conrad T, Kind J, Wirbelauer C, Akhtar A, et al. (2008) Transcription-coupled methylation of histone H3 at lysine 36 regulates dosage compensation by enhancing recruitment of the MSL complex in Drosophila melanogaster. Mol Cell Biol 28: 3401-3409.

156. Scholnick SB, Morgan BA, Hirsh J (1983) The cloned dopa decarboxylase gene is developmentally regulated when reintegrated into the Drosophila genome. Cell 34: 37-45.

157. Spradling AC, Rubin GM (1983) The effect of chromosomal position on the expression of the Drosophila xanthine dehydrogenase gene. Cell 34: 47-57.

158. Sass GL, Pannuti A, Lucchesi JC (2003) Male-specific lethal complex of Drosophila targets activated regions of the X chromosome for chromatin remodeling. Proc Natl Acad Sci U S A 100: 8287-8291.

159. Gelbart ME, Larschan E, Peng S, Park PJ, Kuroda MI (2009) Drosophila MSL complex globally acetylates H4K16 on the male X chromosome for dosage compensation. Nat Struct Mol Biol 16: 825-832.

160. Stenberg P, Pettersson F, Saura AO, Berglund A, Larsson J (2005) Sequence analysis of chromosome identity in three Drosophila species. BMC Bioinformatics 6: 1-17.

161. Gallach M, Arnau V, Marín I (2007) Global patterns of sequence evolution in Drosophila. BMC Genomics 8: 408.

162. Park Y, Mengus G, Bai X, Kageyama Y, Meller VH, et al. (2003) Sequence-specific targeting of Drosophila roX genes by the MSL dosage compensation complex. Mol Cell 11: 977-986.

163. Gilfillan GD, König C, Dahlsveen IK, Prakoura N, Straub T, et al. (2007) Cumulative contributions of weak DNA determinants to targeting the Drosophila dosage compensation complex. Nucleic Acids Res 35: 3561-3572.

164. Legube G, McWeeney SK, Lercher MJ, Akhtar A (2006) X-chromosome-wide profiling of MSL-1 distribution and dosage compensation in Drosophila. Genes Dev 20: 871-883.

165. Philip P, Pettersson F, Stenberg P (2012) Sequence signatures involved in targeting the Male-Specific Lethal complex to X-chromosomal genes in Drosophila melanogaster. BMC Genomics 13: 97.

REFERENCES

70

166. Smith ER, Allis CD, Lucchesi JC (2001) Linking global histone acetylation to the transcription enhancement of X-chromosomal genes in Drosophila males. J Biol Chem 276: 31483-31486.

167. Conrad T, Cavalli FM, Vaquerizas JM, Luscombe NM, Akhtar A (2012) Drosophila dosage compensation involves enhanced Pol II recruitment to male X-linked promoters. Science 337: 742-746.

168. Larschan E, Bishop EP, Kharchenko PV, Core LJ, Lis JT, et al. (2011) X chromosome dosage compensation via enhanced transcriptional elongation in Drosophila. Nature 471: 115-118.

169. Hamada FN, Park PJ, Gordadze PR, Kuroda MI (2005) Global regulation of X chromosomal genes by the MSL complex in Drosophila melanogaster. Genes Dev 19: 2289-2294.

170. Deng X, Koya SK, Kong Y, Meller VH (2009) Coordinated regulation of heterochromatic genes in Drosophila melanogaster males. Genetics 182: 481-491.

171. Zhang Y, Malone JH, Powell SK, Periwal V, Spana E, et al. (2010) Expression in aneuploid Drosophila S2 cells. PLoS Biol 8: e1000320.

172. Stenberg P, Larsson J (2011) Buffering and the evolution of chromosome-wide gene regulation. Chromosoma 120: 213-225.

173. Zhang Y, Oliver B (2010) An evolutionary consequence of dosage compensation on Drosophila melanogaster female X-chromatin structure? BMC Genomics 11: 6.

174. Larsson J, Chen JD, Rasheva V, Rasmuson Lestander A, Pirrotta V (2001) Painting of fourth, a chromosome-specific protein in Drosophila. Proc Natl Acad Sci U S A 98: 6273-6278.

175. Johansson AM, Stenberg P, Allgardsson A, Larsson J (2012) POF regulates the expression of genes on the fourth chromosome in Drosophila melanogaster by binding to nascent RNA. Mol Cell Biol 32: 2121-2134.

176. Locke J, McDermid H (1993) Analysis of Drosophila chromosome four by pulse field electrophoresis. Chromosoma 102: 718-723.

177. Barigozzi C, Dolfini S, Fraccaro M, Raimondi GR, Tiepolo L (1966) In vitro study of the DNA replication patterns of somatic chromosomes of Drosophila melanogaster. Exp Cell Res 43: 231-234.

178. Hochman B (1976) The fourth chromosome of Drosophila melanogaster. In: Ashburner M, Novitski E, editors. The Genetics and biology of Drosophila: Academic Press. pp. 903-928.

179. Miklos GLG, Yamamoto MT, Davies J, Pirrotta V (1988) Microcloning reveals a high frequency of repetitive sequences characteristic of chromosome four and the β-heterochromatin of Drosophila melanogaster. Proc Natl Acad Sci U S A 85: 2051-2055.

180. Sun FL, Cuaycong MH, Craig CA, Wallrath LL, Locke J, et al. (2000) The fourth chromosome of Drosophila melanogaster: interspersed

REFERENCES

71

euchromatic and heterochromatic domains. Proc Natl Acad Sci U S A 97: 5340-5345.

181. Pimpinelli S, Berloco M, Fanti L, Dimitri P, Bonaccorsi S, et al. (1995) Transposable elements are stable structural components of Drosophila melanogaster heterochromatin. Proc Natl Acad Sci U S A 92: 3804-3808.

182. Locke J, Podemski L, Roy K, Pilgrim D, Hodgetts R (1999) Analysis of two cosmid clones from chromosome 4 of Drosophila melanogaster reveals two new genes amid an unusual arrangement of repeated sequences. Genome Res 9: 137-149.

183. Wallrath LL, Elgin SC (1995) Position effect variegation in Drosophila is associated with an altered chromatin structure. Genes Dev 9: 1263-1277.

184. Riddle NC, Elgin SC (2006) The dot chromosome of Drosophila: insights into chromatin states and their change over evolutionary time. Chromosome Res 14: 405-416.

185. Riddle NC, Shaffer CD, Elgin SC (2009) A lot about a little dot - lessons learned from Drosophila melanogaster chromosome 4. Biochem Cell Biol 87: 229-241.

186. Russo CA, Takezaki N, Nei M (1995) Molecular phylogeny and divergence times of drosophilid species. Mol Biol Evol 12: 391-404.

187. Larsson J, Svensson MJ, Stenberg P, Mäkitalo M (2004) Painting of fourth in genus Drosophila suggests autosome-specific gene regulation. Proc Natl Acad Sci U S A 101: 9728-9733.

188. Larsson J, Meller VH (2006) Dosage compensation, the origin and the afterlife of sex chromosomes. Chromosome Res 14: 417-431.

189. Bridges CB (1925) Sex in relation to chromsomes and genes. Am Nat 59: 127-137.

190. Fung STC, Gowen JW (1960) Role of autosome-IV in Drosophila melanogaster sex balance. Genetics 45: 988-989.

191. Sandler L, Novitski E (1956) Evidence for genetic homology between chromosomes I and IV in Drosophila melanogaster, with a proposed explanation for the crowding effect in triploids. Genetics 41: 189-193.

192. FitzPatrick DR (2005) Transcriptional consequences of autosomal trisomy: primary gene dosage with complex downstream effects. Trends Genet 21: 249-253.

193. Brown S (2008) Miscarriage and its associations. Semin Reprod Med 26: 391-400.

194. Rosenbusch B (2004) The incidence of aneuploidy in human oocytes assessed by conventional cytogenetic analysis. Hereditas 141: 97-105.

195. Torres EM, Williams BR, Amon A (2008) Aneuploidy: cells losing their balance. Genetics 179: 737-746.

REFERENCES

72

196. Hassold TJ, Jacobs PA (1984) Trisomy in man. Annu Rev Genet 18: 69-97.

197. Lindsley DL, Sandler L, Baker BS, Carpenter AT, Denell RE, et al. (1972) Segmental aneuploidy and the genetic gross structure of the Drosophila genome. Genetics 71: 157-184.

198. Hodgkin J, Horvitz HR, Brenner S (1979) Nondisjunction mutants of the nematode Caenorhabditis elegans. Genetics 91: 67-94.

199. Stankiewicz P, Lupski JR (2010) Structural variation in the human genome and its role in disease. Annu Rev Med 61: 437-455.

200. Williams BR, Prabhu VR, Hunter KE, Glazier CM, Whittaker CA, et al. (2008) Aneuploidy affects proliferation and spontaneous immortalization in mammalian cells. Science 322: 703-709.

201. Holland AJ, Cleveland DW (2009) Boveri revisited: chromosomal instability, aneuploidy and tumorigenesis. Nat Rev Mol Cell Biol 10: 478-487.

202. Schrider DR, Hahn MW (2010) Gene copy-number polymorphism in nature. Proc Biol Sci 277: 3213-3221.

203. McCarroll SA, Hadnott TN, Perry GH, Sabeti PC, Zody MC, et al. (2006) Common deletion polymorphisms in the human genome. Nat Genet 38: 86-92.

204. Bridges CB (1935) Salivary chromosome maps with a key to the banding of the chromosomes of Drosophila melanogaster. J Hered 26: 60-64.

205. Sorsa V (1988) Chromosome maps of Drosophila. Inc: CRC press. 206. Ashburner M, Golic KG, Hawley RS (2005) Drosophila A laboratory

handbook. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press.

207. Marygold SJ, Roote J, Reuter G, Lambertsson A, Ashburner M, et al. (2007) The ribosomal protein genes and Minute loci of Drosophila melanogaster. Genome Biol 8: R216.

208. Lambertsson A (1998) The Minute genes in Drosophila and their molecular functions. Adv Genet 38: 69-134.

209. Birchler JA, Hiebert JC, Paigen K (1990) Analysis of autosomal dosage compensation involving the alcohol dehydrogenase locus in Drosophila melanogaster. Genetics 124: 679-686.

210. Devlin RH, Holm DG, Grigliatti TA (1982) Autosomal dosage compensation Drosophila melanogaster strains trisomic for the left arm of chromosome 2. Proc Natl Acad Sci U S A 79: 1200-1204.

211. Devlin RH, Holm DG, Grigliatti TA (1988) The influence of whole-arm trisomy on gene expression in Drosophila. Genetics 118: 87-101.

212. Guo M, Birchler JA (1994) Trans-acting dosage effects on the expression of model gene systems in maize aneuploids. Science 266: 1999-2002.

REFERENCES

73

213. Makarevitch I, Phillips RL, Springer NM (2008) Profiling expression changes caused by a segmental aneuploid in maize. BMC Genomics 9: 7.

214. McAnally AA, Yampolsky LY (2010) Widespread transcriptional autosomal dosage compensation in Drosophila correlates with gene expression level. Genome Biol Evol 2: 44-52.

215. FitzPatrick DR, Ramsay J, McGill NI, Shade M, Carothers AD, et al. (2002) Transcriptome analysis of human autosomal trisomy. Hum Mol Genet 11: 3249-3256.

216. Zhang Y, Oliver B (2007) Dosage compensation goes global. Curr Opin Genet Dev 17: 113-120.

217. Torres EM, Dephoure N, Panneerselvam A, Tucker CM, Whittaker CA, et al. (2010) Identification of aneuploidy-tolerating mutations. Cell 143: 71-83.

218. Malone JH, Cho DY, Mattiuzzo NR, Artieri CG, Jiang L, et al. (2012) Mediation of Drosophila autosomal dosage effects and compensation by network interactions. Genome Biol 13: r28.

219. Franke A, Dernburg A, Bashaw GJ, Baker BS (1996) Evidence that MSL-mediated dosage compensation in Drosophila begins at blastoderm. Development 122: 2751-2760.

220. Kind J, Akhtar A (2007) Cotranscriptional recruitment of the dosage compensation complex to X-linked target genes. Genes Dev 21: 2030-2040.

221. Straub T, Grimaud C, Gilfillan GD, Mitterweger A, Becker PB (2008) The chromosomal high-affinity binding sites for the Drosophila dosage compensation complex. PLoS Genet 4: e1000302.

222. Birchler JA, Pal-Bhadra M, Bhadra U (2003) Dosage dependent gene regulation and the compensation of the X chromosome in Drosophila males. Genetica 117: 179-190.

223. Straub T, Gilfillan GD, Maier VK, Becker PB (2005) The Drosophila MSL complex activates the transcription of target genes. Genes Dev 19: 2284-2288.

224. Landfors M, Philip P, Rydén P, Stenberg P (2011) Normalization of high dimensional genomics data where the distribution of the altered variables is skewed. PLoS One 6: e27942.

225. Gorchakov AA, Alekseyenko AA, Kharchenko P, Park PJ, Kuroda MI (2009) Long-range spreading of dosage compensation in Drosophila captures transcribed autosomal genes inserted on X. Genes Dev 23: 2266-2271.

226. Lucchesi JC (1998) Dosage compensation in flies and worms: the ups and downs of X-chromosome regulation. Curr Opin Genet Dev 8: 179-184.

REFERENCES

74

227. Torres EM, Sokolsky T, Tucker CM, Chan LY, Boselli M, et al. (2007) Effects of aneuploidy on cellular physiology and cell division in haploid yeast. Science 317: 916-924.

228. Tang YC, Williams BR, Siegel JJ, Amon A (2011) Identification of aneuploidy-selective antiproliferation compounds. Cell 144: 499-512.

229. Cryderman DE, Vitalini MW, Wallrath LL (2011) Heterochromatin protein 1a is required for an open chromatin structure. Transcription 2: 95-99.

230. Meller VH (2003) Initiation of dosage compensation in Drosophila embryos depends on expression of the roX RNAs. Mech Dev 120: 759-767.

231. de Wit E, Greil F, van Steensel B (2005) Genome-wide HP1 binding in Drosophila: developmental plasticity and genomic targeting signals. Genome Res 15: 1265-1273.

232. Jin Y, Wang Y, Johansen J, Johansen KM (2000) JIL-1, a chromosomal kinase implicated in regulation of chromatin structure, associates with the male specific lethal (MSL) dosage compensation complex. J Cell Biol 149: 1005-1010.

233. Jin Y, Wang Y, Walker DL, Dong H, Conley C, et al. (1999) JIL-1: a novel chromosomal tandem kinase implicated in transcriptional regulation in Drosophila. Mol Cell 4: 129-135.

234. Zhang W, Deng H, Bao X, Lerach S, Girton J, et al. (2006) The JIL-1 histone H3S10 kinase regulates dimethyl H3K9 modifications and heterochromatic spreading in Drosophila. Development 133: 229-235.